Fix async generator type mismatch in Docker Client streaming

- Fixed single_result_generator to properly handle async generators from deep crawl strategies
- Added proper __aiter__ checking to distinguish between CrawlResult and async generators
- Await and yield individual results from nested async generators
- Streaming functionality now works correctly for all patterns (SDK, Direct API, Docker Client)
- All 22 comprehensive tests passing with 100% success rate
- Live streaming test confirmed working end-to-end

.github/FUNDING.yml

@@ -0,0 +1,7 @@
+# These are supported funding model platforms
+
+# GitHub Sponsors
+github: unclecode
+
+# Custom links for enterprise inquiries (uncomment when ready)
+# custom: ["https://crawl4ai.com/enterprise"]

.github/workflows/release.yml

@@ -0,0 +1,142 @@
+name: Release Pipeline
+on:
+  push:
+    tags:
+      - 'v*'
+      - '!test-v*'  # Exclude test tags
+
+jobs:
+  release:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write  # Required for creating releases
+    
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+      
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.12'
+      
+      - name: Extract version from tag
+        id: get_version
+        run: |
+          TAG_VERSION=${GITHUB_REF#refs/tags/v}
+          echo "VERSION=$TAG_VERSION" >> $GITHUB_OUTPUT
+          echo "Releasing version: $TAG_VERSION"
+      
+      - name: Install package dependencies
+        run: |
+          pip install -e .
+      
+      - name: Check version consistency
+        run: |
+          TAG_VERSION=${{ steps.get_version.outputs.VERSION }}
+          PACKAGE_VERSION=$(python -c "from crawl4ai.__version__ import __version__; print(__version__)")
+          
+          echo "Tag version: $TAG_VERSION"
+          echo "Package version: $PACKAGE_VERSION"
+          
+          if [ "$TAG_VERSION" != "$PACKAGE_VERSION" ]; then
+            echo "❌ Version mismatch! Tag: $TAG_VERSION, Package: $PACKAGE_VERSION"
+            echo "Please update crawl4ai/__version__.py to match the tag version"
+            exit 1
+          fi
+          echo "✅ Version check passed: $TAG_VERSION"
+      
+      - name: Install build dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install build twine
+      
+      - name: Build package
+        run: python -m build
+      
+      - name: Check package
+        run: twine check dist/*
+      
+      - name: Upload to PyPI
+        env:
+          TWINE_USERNAME: __token__
+          TWINE_PASSWORD: ${{ secrets.PYPI_TOKEN }}
+        run: |
+          echo "📦 Uploading to PyPI..."
+          twine upload dist/*
+          echo "✅ Package uploaded to https://pypi.org/project/crawl4ai/"
+      
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+      
+      - name: Log in to Docker Hub
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+      
+      - name: Extract major and minor versions
+        id: versions
+        run: |
+          VERSION=${{ steps.get_version.outputs.VERSION }}
+          MAJOR=$(echo $VERSION | cut -d. -f1)
+          MINOR=$(echo $VERSION | cut -d. -f1-2)
+          echo "MAJOR=$MAJOR" >> $GITHUB_OUTPUT
+          echo "MINOR=$MINOR" >> $GITHUB_OUTPUT
+      
+      - name: Build and push Docker images
+        uses: docker/build-push-action@v5
+        with:
+          context: .
+          push: true
+          tags: |
+            unclecode/crawl4ai:${{ steps.get_version.outputs.VERSION }}
+            unclecode/crawl4ai:${{ steps.versions.outputs.MINOR }}
+            unclecode/crawl4ai:${{ steps.versions.outputs.MAJOR }}
+            unclecode/crawl4ai:latest
+          platforms: linux/amd64,linux/arm64
+      
+      - name: Create GitHub Release
+        uses: softprops/action-gh-release@v2
+        with:
+          tag_name: v${{ steps.get_version.outputs.VERSION }}
+          name: Release v${{ steps.get_version.outputs.VERSION }}
+          body: |
+            ## 🎉 Crawl4AI v${{ steps.get_version.outputs.VERSION }} Released!
+            
+            ### 📦 Installation
+            
+            **PyPI:**
+            ```bash
+            pip install crawl4ai==${{ steps.get_version.outputs.VERSION }}
+            ```
+            
+            **Docker:**
+            ```bash
+            docker pull unclecode/crawl4ai:${{ steps.get_version.outputs.VERSION }}
+            docker pull unclecode/crawl4ai:latest
+            ```
+            
+            ### 📝 What's Changed
+            See [CHANGELOG.md](https://github.com/${{ github.repository }}/blob/main/CHANGELOG.md) for details.
+          draft: false
+          prerelease: false
+          token: ${{ secrets.GITHUB_TOKEN }}
+      
+      - name: Summary
+        run: |
+          echo "## 🚀 Release Complete!" >> $GITHUB_STEP_SUMMARY
+          echo "" >> $GITHUB_STEP_SUMMARY
+          echo "### 📦 PyPI Package" >> $GITHUB_STEP_SUMMARY
+          echo "- Version: ${{ steps.get_version.outputs.VERSION }}" >> $GITHUB_STEP_SUMMARY
+          echo "- URL: https://pypi.org/project/crawl4ai/" >> $GITHUB_STEP_SUMMARY
+          echo "- Install: \`pip install crawl4ai==${{ steps.get_version.outputs.VERSION }}\`" >> $GITHUB_STEP_SUMMARY
+          echo "" >> $GITHUB_STEP_SUMMARY
+          echo "### 🐳 Docker Images" >> $GITHUB_STEP_SUMMARY
+          echo "- \`unclecode/crawl4ai:${{ steps.get_version.outputs.VERSION }}\`" >> $GITHUB_STEP_SUMMARY
+          echo "- \`unclecode/crawl4ai:${{ steps.versions.outputs.MINOR }}\`" >> $GITHUB_STEP_SUMMARY
+          echo "- \`unclecode/crawl4ai:${{ steps.versions.outputs.MAJOR }}\`" >> $GITHUB_STEP_SUMMARY
+          echo "- \`unclecode/crawl4ai:latest\`" >> $GITHUB_STEP_SUMMARY
+          echo "" >> $GITHUB_STEP_SUMMARY
+          echo "### 📋 GitHub Release" >> $GITHUB_STEP_SUMMARY
+          echo "https://github.com/${{ github.repository }}/releases/tag/v${{ steps.get_version.outputs.VERSION }}" >> $GITHUB_STEP_SUMMARY

.github/workflows/test-release.yml.disabled

@@ -0,0 +1,116 @@
+name: Test Release Pipeline
+on:
+  push:
+    tags:
+      - 'test-v*'
+
+jobs:
+  test-release:
+    runs-on: ubuntu-latest
+    
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+      
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.12'
+      
+      - name: Extract version from tag
+        id: get_version
+        run: |
+          TAG_VERSION=${GITHUB_REF#refs/tags/test-v}
+          echo "VERSION=$TAG_VERSION" >> $GITHUB_OUTPUT
+          echo "Testing with version: $TAG_VERSION"
+      
+      - name: Install package dependencies
+        run: |
+          pip install -e .
+      
+      - name: Check version consistency
+        run: |
+          TAG_VERSION=${{ steps.get_version.outputs.VERSION }}
+          PACKAGE_VERSION=$(python -c "from crawl4ai.__version__ import __version__; print(__version__)")
+          
+          echo "Tag version: $TAG_VERSION"
+          echo "Package version: $PACKAGE_VERSION"
+          
+          if [ "$TAG_VERSION" != "$PACKAGE_VERSION" ]; then
+            echo "❌ Version mismatch! Tag: $TAG_VERSION, Package: $PACKAGE_VERSION"
+            echo "Please update crawl4ai/__version__.py to match the tag version"
+            exit 1
+          fi
+          echo "✅ Version check passed: $TAG_VERSION"
+      
+      - name: Install build dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install build twine
+      
+      - name: Build package
+        run: python -m build
+      
+      - name: Check package
+        run: twine check dist/*
+      
+      - name: Upload to Test PyPI
+        env:
+          TWINE_USERNAME: __token__
+          TWINE_PASSWORD: ${{ secrets.TEST_PYPI_TOKEN }}
+        run: |
+          echo "📦 Uploading to Test PyPI..."
+          twine upload --repository testpypi dist/* || {
+            if [ $? -eq 1 ]; then
+              echo "⚠️ Upload failed - likely version already exists on Test PyPI"
+              echo "Continuing anyway for test purposes..."
+            else
+              exit 1
+            fi
+          }
+          echo "✅ Test PyPI step complete"
+      
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+      
+      - name: Log in to Docker Hub
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+      
+      - name: Build and push Docker test images
+        uses: docker/build-push-action@v5
+        with:
+          context: .
+          push: true
+          tags: |
+            unclecode/crawl4ai:test-${{ steps.get_version.outputs.VERSION }}
+            unclecode/crawl4ai:test-latest
+          platforms: linux/amd64,linux/arm64
+          cache-from: type=gha
+          cache-to: type=gha,mode=max
+      
+      - name: Summary
+        run: |
+          echo "## 🎉 Test Release Complete!" >> $GITHUB_STEP_SUMMARY
+          echo "" >> $GITHUB_STEP_SUMMARY
+          echo "### 📦 Test PyPI Package" >> $GITHUB_STEP_SUMMARY
+          echo "- Version: ${{ steps.get_version.outputs.VERSION }}" >> $GITHUB_STEP_SUMMARY
+          echo "- URL: https://test.pypi.org/project/crawl4ai/" >> $GITHUB_STEP_SUMMARY
+          echo "- Install: \`pip install -i https://test.pypi.org/simple/ crawl4ai==${{ steps.get_version.outputs.VERSION }}\`" >> $GITHUB_STEP_SUMMARY
+          echo "" >> $GITHUB_STEP_SUMMARY
+          echo "### 🐳 Docker Test Images" >> $GITHUB_STEP_SUMMARY
+          echo "- \`unclecode/crawl4ai:test-${{ steps.get_version.outputs.VERSION }}\`" >> $GITHUB_STEP_SUMMARY
+          echo "- \`unclecode/crawl4ai:test-latest\`" >> $GITHUB_STEP_SUMMARY
+          echo "" >> $GITHUB_STEP_SUMMARY
+          echo "### 🧹 Cleanup Commands" >> $GITHUB_STEP_SUMMARY
+          echo "\`\`\`bash" >> $GITHUB_STEP_SUMMARY
+          echo "# Remove test tag" >> $GITHUB_STEP_SUMMARY
+          echo "git tag -d test-v${{ steps.get_version.outputs.VERSION }}" >> $GITHUB_STEP_SUMMARY
+          echo "git push origin :test-v${{ steps.get_version.outputs.VERSION }}" >> $GITHUB_STEP_SUMMARY
+          echo "" >> $GITHUB_STEP_SUMMARY
+          echo "# Remove Docker test images" >> $GITHUB_STEP_SUMMARY
+          echo "docker rmi unclecode/crawl4ai:test-${{ steps.get_version.outputs.VERSION }}" >> $GITHUB_STEP_SUMMARY
+          echo "docker rmi unclecode/crawl4ai:test-latest" >> $GITHUB_STEP_SUMMARY
+          echo "\`\`\`" >> $GITHUB_STEP_SUMMARY

.gitignore

@@ -1,6 +1,11 @@
 # Scripts folder (private tools)
 .scripts/
 
+# Local development CLI (private)
+local_dev.py
+dev
+DEV_CLI_README.md
+
 # Byte-compiled / optimized / DLL files
 __pycache__/
 *.py[cod]
@@ -270,4 +275,7 @@ docs/**/data
 .codecat/
 
 docs/apps/linkdin/debug*/
-docs/apps/linkdin/samples/insights/*
+docs/apps/linkdin/samples/insights/*
+
+# Production checklist (local, not for version control)
+PRODUCTION_CHECKLIST.md

CHANGELOG.md

@@ -5,6 +5,76 @@ All notable changes to Crawl4AI will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.7.3] - 2025-08-09
+
+### Added
+- **🕵️ Undetected Browser Support**: New browser adapter pattern with stealth capabilities
+  - `browser_adapter.py` with undetected Chrome integration
+  - Bypass sophisticated bot detection systems (Cloudflare, Akamai, custom solutions)
+  - Support for headless stealth mode with anti-detection techniques
+  - Human-like behavior simulation with random mouse movements and scrolling
+  - Comprehensive examples for anti-bot strategies and stealth crawling
+  - Full documentation guide for undetected browser usage
+
+- **🎨 Multi-URL Configuration System**: URL-specific crawler configurations for batch processing
+  - Different crawling strategies for different URL patterns in a single batch
+  - Support for string patterns with wildcards (`"*.pdf"`, `"*/blog/*"`)
+  - Lambda function matchers for complex URL logic
+  - Mixed matchers combining strings and functions with AND/OR logic
+  - Fallback configuration support when no patterns match
+  - First-match-wins configuration selection with optional fallback
+
+- **🧠 Memory Monitoring & Optimization**: Comprehensive memory usage tracking
+  - New `memory_utils.py` module for memory monitoring and optimization
+  - Real-time memory usage tracking during crawl sessions
+  - Memory leak detection and reporting
+  - Performance optimization recommendations
+  - Peak memory usage analysis and efficiency metrics
+  - Automatic cleanup suggestions for memory-intensive operations
+
+- **📊 Enhanced Table Extraction**: Improved table access and DataFrame conversion
+  - Direct `result.tables` interface replacing generic `result.media` approach
+  - Instant pandas DataFrame conversion with `pd.DataFrame(table['data'])`
+  - Enhanced table detection algorithms for better accuracy
+  - Table metadata including source XPath and headers
+  - Improved table structure preservation during extraction
+
+- **💰 GitHub Sponsors Integration**: 4-tier sponsorship system
+  - Supporter ($5/month): Community support + early feature previews
+  - Professional ($25/month): Priority support + beta access
+  - Business ($100/month): Direct consultation + custom integrations
+  - Enterprise ($500/month): Dedicated support + feature development
+  - Custom arrangement options for larger organizations
+
+- **🐳 Docker LLM Provider Flexibility**: Environment-based LLM configuration
+  - `LLM_PROVIDER` environment variable support for dynamic provider switching
+  - `.llm.env` file support for secure configuration management
+  - Per-request provider override capabilities in API endpoints
+  - Support for OpenAI, Groq, and other providers without rebuilding images
+  - Enhanced Docker documentation with deployment examples
+
+### Fixed
+- **URL Matcher Fallback**: Resolved edge cases in URL pattern matching logic
+- **Memory Management**: Fixed memory leaks in long-running crawl sessions
+- **Sitemap Processing**: Improved redirect handling in sitemap fetching
+- **Table Extraction**: Enhanced table detection and extraction accuracy
+- **Error Handling**: Better error messages and recovery from network failures
+
+### Changed
+- **Architecture Refactoring**: Major cleanup and optimization
+  - Moved 2,450+ lines from main `async_crawler_strategy.py` to backup
+  - Cleaner separation of concerns in crawler architecture
+  - Better maintainability and code organization
+  - Preserved backward compatibility while improving performance
+
+### Documentation
+- **Comprehensive Examples**: Added real-world URLs and practical use cases
+- **API Documentation**: Complete CrawlResult field documentation with all available fields
+- **Migration Guides**: Updated table extraction patterns from `result.media` to `result.tables`
+- **Undetected Browser Guide**: Full documentation for stealth mode and anti-bot strategies
+- **Multi-Config Examples**: Detailed examples for URL-specific configurations
+- **Docker Deployment**: Enhanced Docker documentation with LLM provider configuration
+
 ## [0.7.x] - 2025-06-29
 
 ### Added
@@ -21,6 +91,21 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+### Added
+- **Flexible LLM Provider Configuration** (Docker): 
+  - Support for `LLM_PROVIDER` environment variable to override default provider
+  - Per-request provider override via optional `provider` parameter in API endpoints
+  - Automatic provider validation with clear error messages
+  - Updated Docker documentation and examples
+
+### Changed
+- **WebScrapingStrategy Refactoring**: Simplified content scraping architecture
+  - `WebScrapingStrategy` is now an alias for `LXMLWebScrapingStrategy` for backward compatibility
+  - Removed redundant BeautifulSoup-based implementation (~1000 lines of code)
+  - `LXMLWebScrapingStrategy` now inherits directly from `ContentScrapingStrategy`
+  - All existing code using `WebScrapingStrategy` continues to work without modification
+  - Default scraping strategy remains `LXMLWebScrapingStrategy` for optimal performance
+
 ### Added
 - **AsyncUrlSeeder**: High-performance URL discovery system for intelligent crawling at scale
   - Discover URLs from sitemaps and Common Crawl index

README-first.md

@@ -0,0 +1,809 @@
+# 🚀🤖 Crawl4AI: Open-source LLM Friendly Web Crawler & Scraper.
+
+<div align="center">
+
+<a href="https://trendshift.io/repositories/11716" target="_blank"><img src="https://trendshift.io/api/badge/repositories/11716" alt="unclecode%2Fcrawl4ai | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a>
+
+[![GitHub Stars](https://img.shields.io/github/stars/unclecode/crawl4ai?style=social)](https://github.com/unclecode/crawl4ai/stargazers)
+[![GitHub Forks](https://img.shields.io/github/forks/unclecode/crawl4ai?style=social)](https://github.com/unclecode/crawl4ai/network/members)
+
+[![PyPI version](https://badge.fury.io/py/crawl4ai.svg)](https://badge.fury.io/py/crawl4ai)
+[![Python Version](https://img.shields.io/pypi/pyversions/crawl4ai)](https://pypi.org/project/crawl4ai/)
+[![Downloads](https://static.pepy.tech/badge/crawl4ai/month)](https://pepy.tech/project/crawl4ai)
+[![GitHub Sponsors](https://img.shields.io/github/sponsors/unclecode?style=flat&logo=GitHub-Sponsors&label=Sponsors&color=pink)](https://github.com/sponsors/unclecode)
+
+<p align="center">
+    <a href="https://x.com/crawl4ai">
+      <img src="https://img.shields.io/badge/Follow%20on%20X-000000?style=for-the-badge&logo=x&logoColor=white" alt="Follow on X" />
+    </a>
+    <a href="https://www.linkedin.com/company/crawl4ai">
+      <img src="https://img.shields.io/badge/Follow%20on%20LinkedIn-0077B5?style=for-the-badge&logo=linkedin&logoColor=white" alt="Follow on LinkedIn" />
+    </a>
+    <a href="https://discord.gg/jP8KfhDhyN">
+      <img src="https://img.shields.io/badge/Join%20our%20Discord-5865F2?style=for-the-badge&logo=discord&logoColor=white" alt="Join our Discord" />
+    </a>
+  </p>
+</div>
+
+Crawl4AI is the #1 trending GitHub repository, actively maintained by a vibrant community. It delivers blazing-fast, AI-ready web crawling tailored for LLMs, AI agents, and data pipelines. Open source, flexible, and built for real-time performance, Crawl4AI empowers developers with unmatched speed, precision, and deployment ease.  
+
+[✨ Check out latest update v0.7.0](#-recent-updates)
+
+🎉 **Version 0.7.0 is now available!** The Adaptive Intelligence Update introduces groundbreaking features: Adaptive Crawling that learns website patterns, Virtual Scroll support for infinite pages, intelligent Link Preview with 3-layer scoring, Async URL Seeder for massive discovery, and significant performance improvements. [Read the release notes →](https://github.com/unclecode/crawl4ai/blob/main/docs/blog/release-v0.7.0.md)
+
+<details>
+<summary>🤓 <strong>My Personal Story</strong></summary>
+
+My journey with computers started in childhood when my dad, a computer scientist, introduced me to an Amstrad computer. Those early days sparked a fascination with technology, leading me to pursue computer science and specialize in NLP during my postgraduate studies. It was during this time that I first delved into web crawling, building tools to help researchers organize papers and extract information from publications a challenging yet rewarding experience that honed my skills in data extraction.
+
+Fast forward to 2023, I was working on a tool for a project and needed a crawler to convert a webpage into markdown. While exploring solutions, I found one that claimed to be open-source but required creating an account and generating an API token. Worse, it turned out to be a SaaS model charging $16, and its quality didn’t meet my standards. Frustrated, I realized this was a deeper problem. That frustration turned into turbo anger mode, and I decided to build my own solution. In just a few days, I created Crawl4AI. To my surprise, it went viral, earning thousands of GitHub stars and resonating with a global community.
+
+I made Crawl4AI open-source for two reasons. First, it’s my way of giving back to the open-source community that has supported me throughout my career. Second, I believe data should be accessible to everyone, not locked behind paywalls or monopolized by a few. Open access to data lays the foundation for the democratization of AI, a vision where individuals can train their own models and take ownership of their information. This library is the first step in a larger journey to create the best open-source data extraction and generation tool the world has ever seen, built collaboratively by a passionate community.
+
+Thank you to everyone who has supported this project, used it, and shared feedback. Your encouragement motivates me to dream even bigger. Join us, file issues, submit PRs, or spread the word. Together, we can build a tool that truly empowers people to access their own data and reshape the future of AI.
+</details>
+
+## 🧐 Why Crawl4AI?
+
+1. **Built for LLMs**: Creates smart, concise Markdown optimized for RAG and fine-tuning applications.  
+2. **Lightning Fast**: Delivers results faster with real-time, cost-efficient performance.  
+3. **Flexible Browser Control**: Offers session management, proxies, and custom hooks for seamless data access.  
+4. **Heuristic Intelligence**: Uses advanced algorithms for efficient extraction, reducing reliance on costly models.  
+5. **Open Source & Deployable**: Fully open-source with no API keys—ready for Docker and cloud integration.  
+6. **Thriving Community**: Actively maintained by a vibrant community and the #1 trending GitHub repository.
+
+## 🚀 Quick Start 
+
+1. Install Crawl4AI:
+```bash
+# Install the package
+pip install -U crawl4ai
+
+# For pre release versions
+pip install crawl4ai --pre
+
+# Run post-installation setup
+crawl4ai-setup
+
+# Verify your installation
+crawl4ai-doctor
+```
+
+If you encounter any browser-related issues, you can install them manually:
+```bash
+python -m playwright install --with-deps chromium
+```
+
+2. Run a simple web crawl with Python:
+```python
+import asyncio
+from crawl4ai import *
+
+async def main():
+    async with AsyncWebCrawler() as crawler:
+        result = await crawler.arun(
+            url="https://www.nbcnews.com/business",
+        )
+        print(result.markdown)
+
+if __name__ == "__main__":
+    asyncio.run(main())
+```
+
+3. Or use the new command-line interface:
+```bash
+# Basic crawl with markdown output
+crwl https://www.nbcnews.com/business -o markdown
+
+# Deep crawl with BFS strategy, max 10 pages
+crwl https://docs.crawl4ai.com --deep-crawl bfs --max-pages 10
+
+# Use LLM extraction with a specific question
+crwl https://www.example.com/products -q "Extract all product prices"
+```
+
+## ✨ Features 
+
+<details>
+<summary>📝 <strong>Markdown Generation</strong></summary>
+
+- 🧹 **Clean Markdown**: Generates clean, structured Markdown with accurate formatting.
+- 🎯 **Fit Markdown**: Heuristic-based filtering to remove noise and irrelevant parts for AI-friendly processing.
+- 🔗 **Citations and References**: Converts page links into a numbered reference list with clean citations.
+- 🛠️ **Custom Strategies**: Users can create their own Markdown generation strategies tailored to specific needs.
+- 📚 **BM25 Algorithm**: Employs BM25-based filtering for extracting core information and removing irrelevant content. 
+</details>
+
+<details>
+<summary>📊 <strong>Structured Data Extraction</strong></summary>
+
+- 🤖 **LLM-Driven Extraction**: Supports all LLMs (open-source and proprietary) for structured data extraction.
+- 🧱 **Chunking Strategies**: Implements chunking (topic-based, regex, sentence-level) for targeted content processing.
+- 🌌 **Cosine Similarity**: Find relevant content chunks based on user queries for semantic extraction.
+- 🔎 **CSS-Based Extraction**: Fast schema-based data extraction using XPath and CSS selectors.
+- 🔧 **Schema Definition**: Define custom schemas for extracting structured JSON from repetitive patterns.
+
+</details>
+
+<details>
+<summary>🌐 <strong>Browser Integration</strong></summary>
+
+- 🖥️ **Managed Browser**: Use user-owned browsers with full control, avoiding bot detection.
+- 🔄 **Remote Browser Control**: Connect to Chrome Developer Tools Protocol for remote, large-scale data extraction.
+- 👤 **Browser Profiler**: Create and manage persistent profiles with saved authentication states, cookies, and settings.
+- 🔒 **Session Management**: Preserve browser states and reuse them for multi-step crawling.
+- 🧩 **Proxy Support**: Seamlessly connect to proxies with authentication for secure access.
+- ⚙️ **Full Browser Control**: Modify headers, cookies, user agents, and more for tailored crawling setups.
+- 🌍 **Multi-Browser Support**: Compatible with Chromium, Firefox, and WebKit.
+- 📐 **Dynamic Viewport Adjustment**: Automatically adjusts the browser viewport to match page content, ensuring complete rendering and capturing of all elements.
+
+</details>
+
+<details>
+<summary>🔎 <strong>Crawling & Scraping</strong></summary>
+
+- 🖼️ **Media Support**: Extract images, audio, videos, and responsive image formats like `srcset` and `picture`.
+- 🚀 **Dynamic Crawling**: Execute JS and wait for async or sync for dynamic content extraction.
+- 📸 **Screenshots**: Capture page screenshots during crawling for debugging or analysis.
+- 📂 **Raw Data Crawling**: Directly process raw HTML (`raw:`) or local files (`file://`).
+- 🔗 **Comprehensive Link Extraction**: Extracts internal, external links, and embedded iframe content.
+- 🛠️ **Customizable Hooks**: Define hooks at every step to customize crawling behavior.
+- 💾 **Caching**: Cache data for improved speed and to avoid redundant fetches.
+- 📄 **Metadata Extraction**: Retrieve structured metadata from web pages.
+- 📡 **IFrame Content Extraction**: Seamless extraction from embedded iframe content.
+- 🕵️ **Lazy Load Handling**: Waits for images to fully load, ensuring no content is missed due to lazy loading.
+- 🔄 **Full-Page Scanning**: Simulates scrolling to load and capture all dynamic content, perfect for infinite scroll pages.
+
+</details>
+
+<details>
+<summary>🚀 <strong>Deployment</strong></summary>
+
+- 🐳 **Dockerized Setup**: Optimized Docker image with FastAPI server for easy deployment.
+- 🔑 **Secure Authentication**: Built-in JWT token authentication for API security.
+- 🔄 **API Gateway**: One-click deployment with secure token authentication for API-based workflows.
+- 🌐 **Scalable Architecture**: Designed for mass-scale production and optimized server performance.
+- ☁️ **Cloud Deployment**: Ready-to-deploy configurations for major cloud platforms.
+
+</details>
+
+<details>
+<summary>🎯 <strong>Additional Features</strong></summary>
+
+- 🕶️ **Stealth Mode**: Avoid bot detection by mimicking real users.
+- 🏷️ **Tag-Based Content Extraction**: Refine crawling based on custom tags, headers, or metadata.
+- 🔗 **Link Analysis**: Extract and analyze all links for detailed data exploration.
+- 🛡️ **Error Handling**: Robust error management for seamless execution.
+- 🔐 **CORS & Static Serving**: Supports filesystem-based caching and cross-origin requests.
+- 📖 **Clear Documentation**: Simplified and updated guides for onboarding and advanced usage.
+- 🙌 **Community Recognition**: Acknowledges contributors and pull requests for transparency.
+
+</details>
+
+## Try it Now!
+
+✨ Play around with this [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1SgRPrByQLzjRfwoRNq1wSGE9nYY_EE8C?usp=sharing)
+
+✨ Visit our [Documentation Website](https://docs.crawl4ai.com/)
+
+## Installation 🛠️
+
+Crawl4AI offers flexible installation options to suit various use cases. You can install it as a Python package or use Docker.
+
+<details>
+<summary>🐍 <strong>Using pip</strong></summary>
+
+Choose the installation option that best fits your needs:
+
+### Basic Installation
+
+For basic web crawling and scraping tasks:
+
+```bash
+pip install crawl4ai
+crawl4ai-setup # Setup the browser
+```
+
+By default, this will install the asynchronous version of Crawl4AI, using Playwright for web crawling.
+
+👉 **Note**: When you install Crawl4AI, the `crawl4ai-setup` should automatically install and set up Playwright. However, if you encounter any Playwright-related errors, you can manually install it using one of these methods:
+
+1. Through the command line:
+
+   ```bash
+   playwright install
+   ```
+
+2. If the above doesn't work, try this more specific command:
+
+   ```bash
+   python -m playwright install chromium
+   ```
+
+This second method has proven to be more reliable in some cases.
+
+---
+
+### Installation with Synchronous Version
+
+The sync version is deprecated and will be removed in future versions. If you need the synchronous version using Selenium:
+
+```bash
+pip install crawl4ai[sync]
+```
+
+---
+
+### Development Installation
+
+For contributors who plan to modify the source code:
+
+```bash
+git clone https://github.com/unclecode/crawl4ai.git
+cd crawl4ai
+pip install -e .                    # Basic installation in editable mode
+```
+
+Install optional features:
+
+```bash
+pip install -e ".[torch]"           # With PyTorch features
+pip install -e ".[transformer]"     # With Transformer features
+pip install -e ".[cosine]"          # With cosine similarity features
+pip install -e ".[sync]"            # With synchronous crawling (Selenium)
+pip install -e ".[all]"             # Install all optional features
+```
+
+</details>
+
+<details>
+<summary>🐳 <strong>Docker Deployment</strong></summary>
+
+> 🚀 **Now Available!** Our completely redesigned Docker implementation is here! This new solution makes deployment more efficient and seamless than ever.
+
+### New Docker Features
+
+The new Docker implementation includes:
+- **Browser pooling** with page pre-warming for faster response times
+- **Interactive playground** to test and generate request code
+- **MCP integration** for direct connection to AI tools like Claude Code
+- **Comprehensive API endpoints** including HTML extraction, screenshots, PDF generation, and JavaScript execution
+- **Multi-architecture support** with automatic detection (AMD64/ARM64)
+- **Optimized resources** with improved memory management
+
+### Getting Started
+
+```bash
+# Pull and run the latest release candidate
+docker pull unclecode/crawl4ai:0.7.0
+docker run -d -p 11235:11235 --name crawl4ai --shm-size=1g unclecode/crawl4ai:0.7.0
+
+# Visit the playground at http://localhost:11235/playground
+```
+
+For complete documentation, see our [Docker Deployment Guide](https://docs.crawl4ai.com/core/docker-deployment/).
+
+</details>
+
+---
+
+### Quick Test
+
+Run a quick test (works for both Docker options):
+
+```python
+import requests
+
+# Submit a crawl job
+response = requests.post(
+    "http://localhost:11235/crawl",
+    json={"urls": ["https://example.com"], "priority": 10}
+)
+if response.status_code == 200:
+    print("Crawl job submitted successfully.")
+    
+if "results" in response.json():
+    results = response.json()["results"]
+    print("Crawl job completed. Results:")
+    for result in results:
+        print(result)
+else:
+    task_id = response.json()["task_id"]
+    print(f"Crawl job submitted. Task ID:: {task_id}")
+    result = requests.get(f"http://localhost:11235/task/{task_id}")
+```
+
+For more examples, see our [Docker Examples](https://github.com/unclecode/crawl4ai/blob/main/docs/examples/docker_example.py). For advanced configuration, environment variables, and usage examples, see our [Docker Deployment Guide](https://docs.crawl4ai.com/basic/docker-deployment/).
+
+</details>
+
+
+## 🔬 Advanced Usage Examples 🔬
+
+You can check the project structure in the directory [docs/examples](https://github.com/unclecode/crawl4ai/tree/main/docs/examples). Over there, you can find a variety of examples; here, some popular examples are shared.
+
+<details>
+<summary>📝 <strong>Heuristic Markdown Generation with Clean and Fit Markdown</strong></summary>
+
+```python
+import asyncio
+from crawl4ai import AsyncWebCrawler, BrowserConfig, CrawlerRunConfig, CacheMode
+from crawl4ai.content_filter_strategy import PruningContentFilter, BM25ContentFilter
+from crawl4ai.markdown_generation_strategy import DefaultMarkdownGenerator
+
+async def main():
+    browser_config = BrowserConfig(
+        headless=True,  
+        verbose=True,
+    )
+    run_config = CrawlerRunConfig(
+        cache_mode=CacheMode.ENABLED,
+        markdown_generator=DefaultMarkdownGenerator(
+            content_filter=PruningContentFilter(threshold=0.48, threshold_type="fixed", min_word_threshold=0)
+        ),
+        # markdown_generator=DefaultMarkdownGenerator(
+        #     content_filter=BM25ContentFilter(user_query="WHEN_WE_FOCUS_BASED_ON_A_USER_QUERY", bm25_threshold=1.0)
+        # ),
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        result = await crawler.arun(
+            url="https://docs.micronaut.io/4.7.6/guide/",
+            config=run_config
+        )
+        print(len(result.markdown.raw_markdown))
+        print(len(result.markdown.fit_markdown))
+
+if __name__ == "__main__":
+    asyncio.run(main())
+```
+
+</details>
+
+<details>
+<summary>🖥️ <strong>Executing JavaScript & Extract Structured Data without LLMs</strong></summary>
+
+```python
+import asyncio
+from crawl4ai import AsyncWebCrawler, BrowserConfig, CrawlerRunConfig, CacheMode
+from crawl4ai import JsonCssExtractionStrategy
+import json
+
+async def main():
+    schema = {
+    "name": "KidoCode Courses",
+    "baseSelector": "section.charge-methodology .w-tab-content > div",
+    "fields": [
+        {
+            "name": "section_title",
+            "selector": "h3.heading-50",
+            "type": "text",
+        },
+        {
+            "name": "section_description",
+            "selector": ".charge-content",
+            "type": "text",
+        },
+        {
+            "name": "course_name",
+            "selector": ".text-block-93",
+            "type": "text",
+        },
+        {
+            "name": "course_description",
+            "selector": ".course-content-text",
+            "type": "text",
+        },
+        {
+            "name": "course_icon",
+            "selector": ".image-92",
+            "type": "attribute",
+            "attribute": "src"
+        }
+    }
+}
+
+    extraction_strategy = JsonCssExtractionStrategy(schema, verbose=True)
+
+    browser_config = BrowserConfig(
+        headless=False,
+        verbose=True
+    )
+    run_config = CrawlerRunConfig(
+        extraction_strategy=extraction_strategy,
+        js_code=["""(async () => {const tabs = document.querySelectorAll("section.charge-methodology .tabs-menu-3 > div");for(let tab of tabs) {tab.scrollIntoView();tab.click();await new Promise(r => setTimeout(r, 500));}})();"""],
+        cache_mode=CacheMode.BYPASS
+    )
+        
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        
+        result = await crawler.arun(
+            url="https://www.kidocode.com/degrees/technology",
+            config=run_config
+        )
+
+        companies = json.loads(result.extracted_content)
+        print(f"Successfully extracted {len(companies)} companies")
+        print(json.dumps(companies[0], indent=2))
+
+
+if __name__ == "__main__":
+    asyncio.run(main())
+```
+
+</details>
+
+<details>
+<summary>📚 <strong>Extracting Structured Data with LLMs</strong></summary>
+
+```python
+import os
+import asyncio
+from crawl4ai import AsyncWebCrawler, BrowserConfig, CrawlerRunConfig, CacheMode, LLMConfig
+from crawl4ai import LLMExtractionStrategy
+from pydantic import BaseModel, Field
+
+class OpenAIModelFee(BaseModel):
+    model_name: str = Field(..., description="Name of the OpenAI model.")
+    input_fee: str = Field(..., description="Fee for input token for the OpenAI model.")
+    output_fee: str = Field(..., description="Fee for output token for the OpenAI model.")
+
+async def main():
+    browser_config = BrowserConfig(verbose=True)
+    run_config = CrawlerRunConfig(
+        word_count_threshold=1,
+        extraction_strategy=LLMExtractionStrategy(
+            # Here you can use any provider that Litellm library supports, for instance: ollama/qwen2
+            # provider="ollama/qwen2", api_token="no-token", 
+            llm_config = LLMConfig(provider="openai/gpt-4o", api_token=os.getenv('OPENAI_API_KEY')), 
+            schema=OpenAIModelFee.schema(),
+            extraction_type="schema",
+            instruction="""From the crawled content, extract all mentioned model names along with their fees for input and output tokens. 
+            Do not miss any models in the entire content. One extracted model JSON format should look like this: 
+            {"model_name": "GPT-4", "input_fee": "US$10.00 / 1M tokens", "output_fee": "US$30.00 / 1M tokens"}."""
+        ),            
+        cache_mode=CacheMode.BYPASS,
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        result = await crawler.arun(
+            url='https://openai.com/api/pricing/',
+            config=run_config
+        )
+        print(result.extracted_content)
+
+if __name__ == "__main__":
+    asyncio.run(main())
+```
+
+</details>
+
+<details>
+<summary>🤖 <strong>Using Your own Browser with Custom User Profile</strong></summary>
+
+```python
+import os, sys
+from pathlib import Path
+import asyncio, time
+from crawl4ai import AsyncWebCrawler, BrowserConfig, CrawlerRunConfig, CacheMode
+
+async def test_news_crawl():
+    # Create a persistent user data directory
+    user_data_dir = os.path.join(Path.home(), ".crawl4ai", "browser_profile")
+    os.makedirs(user_data_dir, exist_ok=True)
+
+    browser_config = BrowserConfig(
+        verbose=True,
+        headless=True,
+        user_data_dir=user_data_dir,
+        use_persistent_context=True,
+    )
+    run_config = CrawlerRunConfig(
+        cache_mode=CacheMode.BYPASS
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        url = "ADDRESS_OF_A_CHALLENGING_WEBSITE"
+        
+        result = await crawler.arun(
+            url,
+            config=run_config,
+            magic=True,
+        )
+        
+        print(f"Successfully crawled {url}")
+        print(f"Content length: {len(result.markdown)}")
+```
+
+</details>
+
+## ✨ Recent Updates
+
+### Version 0.7.0 Release Highlights - The Adaptive Intelligence Update
+
+- **🧠 Adaptive Crawling**: Your crawler now learns and adapts to website patterns automatically:
+  ```python
+  config = AdaptiveConfig(
+      confidence_threshold=0.7, # Min confidence to stop crawling
+      max_depth=5, # Maximum crawl depth
+      max_pages=20, # Maximum number of pages to crawl
+      strategy="statistical"
+  )
+  
+  async with AsyncWebCrawler() as crawler:
+      adaptive_crawler = AdaptiveCrawler(crawler, config)
+      state = await adaptive_crawler.digest(
+          start_url="https://news.example.com",
+          query="latest news content"
+      )
+  # Crawler learns patterns and improves extraction over time
+  ```
+
+- **🌊 Virtual Scroll Support**: Complete content extraction from infinite scroll pages:
+  ```python
+  scroll_config = VirtualScrollConfig(
+      container_selector="[data-testid='feed']",
+      scroll_count=20,
+      scroll_by="container_height",
+      wait_after_scroll=1.0
+  )
+  
+  result = await crawler.arun(url, config=CrawlerRunConfig(
+      virtual_scroll_config=scroll_config
+  ))
+  ```
+
+- **🔗 Intelligent Link Analysis**: 3-layer scoring system for smart link prioritization:
+  ```python
+  link_config = LinkPreviewConfig(
+      query="machine learning tutorials",
+      score_threshold=0.3,
+      concurrent_requests=10
+  )
+  
+  result = await crawler.arun(url, config=CrawlerRunConfig(
+      link_preview_config=link_config,
+      score_links=True
+  ))
+  # Links ranked by relevance and quality
+  ```
+
+- **🎣 Async URL Seeder**: Discover thousands of URLs in seconds:
+  ```python
+  seeder = AsyncUrlSeeder(SeedingConfig(
+      source="sitemap+cc",
+      pattern="*/blog/*",
+      query="python tutorials",
+      score_threshold=0.4
+  ))
+  
+  urls = await seeder.discover("https://example.com")
+  ```
+
+- **⚡ Performance Boost**: Up to 3x faster with optimized resource handling and memory efficiency
+
+Read the full details in our [0.7.0 Release Notes](https://docs.crawl4ai.com/blog/release-v0.7.0) or check the [CHANGELOG](https://github.com/unclecode/crawl4ai/blob/main/CHANGELOG.md).
+
+## Version Numbering in Crawl4AI
+
+Crawl4AI follows standard Python version numbering conventions (PEP 440) to help users understand the stability and features of each release.
+
+### Version Numbers Explained
+
+Our version numbers follow this pattern: `MAJOR.MINOR.PATCH` (e.g., 0.4.3)
+
+#### Pre-release Versions
+We use different suffixes to indicate development stages:
+
+- `dev` (0.4.3dev1): Development versions, unstable
+- `a` (0.4.3a1): Alpha releases, experimental features
+- `b` (0.4.3b1): Beta releases, feature complete but needs testing
+- `rc` (0.4.3): Release candidates, potential final version
+
+#### Installation
+- Regular installation (stable version):
+  ```bash
+  pip install -U crawl4ai
+  ```
+
+- Install pre-release versions:
+  ```bash
+  pip install crawl4ai --pre
+  ```
+
+- Install specific version:
+  ```bash
+  pip install crawl4ai==0.4.3b1
+  ```
+
+#### Why Pre-releases?
+We use pre-releases to:
+- Test new features in real-world scenarios
+- Gather feedback before final releases
+- Ensure stability for production users
+- Allow early adopters to try new features
+
+For production environments, we recommend using the stable version. For testing new features, you can opt-in to pre-releases using the `--pre` flag.
+
+## 📖 Documentation & Roadmap 
+
+> 🚨 **Documentation Update Alert**: We're undertaking a major documentation overhaul next week to reflect recent updates and improvements. Stay tuned for a more comprehensive and up-to-date guide!
+
+For current documentation, including installation instructions, advanced features, and API reference, visit our [Documentation Website](https://docs.crawl4ai.com/).
+
+To check our development plans and upcoming features, visit our [Roadmap](https://github.com/unclecode/crawl4ai/blob/main/ROADMAP.md).
+
+<details>
+<summary>📈 <strong>Development TODOs</strong></summary>
+
+- [x] 0. Graph Crawler: Smart website traversal using graph search algorithms for comprehensive nested page extraction
+- [ ] 1. Question-Based Crawler: Natural language driven web discovery and content extraction
+- [ ] 2. Knowledge-Optimal Crawler: Smart crawling that maximizes knowledge while minimizing data extraction
+- [ ] 3. Agentic Crawler: Autonomous system for complex multi-step crawling operations
+- [ ] 4. Automated Schema Generator: Convert natural language to extraction schemas
+- [ ] 5. Domain-Specific Scrapers: Pre-configured extractors for common platforms (academic, e-commerce)
+- [ ] 6. Web Embedding Index: Semantic search infrastructure for crawled content
+- [ ] 7. Interactive Playground: Web UI for testing, comparing strategies with AI assistance
+- [ ] 8. Performance Monitor: Real-time insights into crawler operations
+- [ ] 9. Cloud Integration: One-click deployment solutions across cloud providers
+- [ ] 10. Sponsorship Program: Structured support system with tiered benefits
+- [ ] 11. Educational Content: "How to Crawl" video series and interactive tutorials
+
+</details>
+
+## 🤝 Contributing 
+
+We welcome contributions from the open-source community. Check out our [contribution guidelines](https://github.com/unclecode/crawl4ai/blob/main/CONTRIBUTORS.md) for more information.
+
+I'll help modify the license section with badges. For the halftone effect, here's a version with it:
+
+Here's the updated license section:
+
+## 📄 License & Attribution
+
+This project is licensed under the Apache License 2.0, attribution is recommended via the badges below. See the [Apache 2.0 License](https://github.com/unclecode/crawl4ai/blob/main/LICENSE) file for details.
+
+### Attribution Requirements
+When using Crawl4AI, you must include one of the following attribution methods:
+
+#### 1. Badge Attribution (Recommended)
+Add one of these badges to your README, documentation, or website:
+
+| Theme | Badge |
+|-------|-------|
+| **Disco Theme (Animated)** | <a href="https://github.com/unclecode/crawl4ai"><img src="./docs/assets/powered-by-disco.svg" alt="Powered by Crawl4AI" width="200"/></a> |
+| **Night Theme (Dark with Neon)** | <a href="https://github.com/unclecode/crawl4ai"><img src="./docs/assets/powered-by-night.svg" alt="Powered by Crawl4AI" width="200"/></a> |
+| **Dark Theme (Classic)** | <a href="https://github.com/unclecode/crawl4ai"><img src="./docs/assets/powered-by-dark.svg" alt="Powered by Crawl4AI" width="200"/></a> |
+| **Light Theme (Classic)** | <a href="https://github.com/unclecode/crawl4ai"><img src="./docs/assets/powered-by-light.svg" alt="Powered by Crawl4AI" width="200"/></a> |
+ 
+
+HTML code for adding the badges:
+```html
+<!-- Disco Theme (Animated) -->
+<a href="https://github.com/unclecode/crawl4ai">
+  <img src="https://raw.githubusercontent.com/unclecode/crawl4ai/main/docs/assets/powered-by-disco.svg" alt="Powered by Crawl4AI" width="200"/>
+</a>
+
+<!-- Night Theme (Dark with Neon) -->
+<a href="https://github.com/unclecode/crawl4ai">
+  <img src="https://raw.githubusercontent.com/unclecode/crawl4ai/main/docs/assets/powered-by-night.svg" alt="Powered by Crawl4AI" width="200"/>
+</a>
+
+<!-- Dark Theme (Classic) -->
+<a href="https://github.com/unclecode/crawl4ai">
+  <img src="https://raw.githubusercontent.com/unclecode/crawl4ai/main/docs/assets/powered-by-dark.svg" alt="Powered by Crawl4AI" width="200"/>
+</a>
+
+<!-- Light Theme (Classic) -->
+<a href="https://github.com/unclecode/crawl4ai">
+  <img src="https://raw.githubusercontent.com/unclecode/crawl4ai/main/docs/assets/powered-by-light.svg" alt="Powered by Crawl4AI" width="200"/>
+</a>
+
+<!-- Simple Shield Badge -->
+<a href="https://github.com/unclecode/crawl4ai">
+  <img src="https://img.shields.io/badge/Powered%20by-Crawl4AI-blue?style=flat-square" alt="Powered by Crawl4AI"/>
+</a>
+```
+
+#### 2. Text Attribution
+Add this line to your documentation:
+```
+This project uses Crawl4AI (https://github.com/unclecode/crawl4ai) for web data extraction.
+```
+
+## 📚 Citation
+
+If you use Crawl4AI in your research or project, please cite:
+
+```bibtex
+@software{crawl4ai2024,
+  author = {UncleCode},
+  title = {Crawl4AI: Open-source LLM Friendly Web Crawler & Scraper},
+  year = {2024},
+  publisher = {GitHub},
+  journal = {GitHub Repository},
+  howpublished = {\url{https://github.com/unclecode/crawl4ai}},
+  commit = {Please use the commit hash you're working with}
+}
+```
+
+Text citation format:
+```
+UncleCode. (2024). Crawl4AI: Open-source LLM Friendly Web Crawler & Scraper [Computer software]. 
+GitHub. https://github.com/unclecode/crawl4ai
+```
+
+## 📧 Contact 
+
+For questions, suggestions, or feedback, feel free to reach out:
+
+- GitHub: [unclecode](https://github.com/unclecode)
+- Twitter: [@unclecode](https://twitter.com/unclecode)
+- Website: [crawl4ai.com](https://crawl4ai.com)
+
+Happy Crawling! 🕸️🚀
+
+## 💖 Support Crawl4AI
+
+> 🎉 **Sponsorship Program Just Launched!** Be among the first 50 **Founding Sponsors** and get permanent recognition in our Hall of Fame!
+
+Crawl4AI is the #1 trending open-source web crawler with 51K+ stars. Your support ensures we stay independent, innovative, and free forever.
+
+<div align="center">
+
+[![Become a Sponsor](https://img.shields.io/badge/Become%20a%20Sponsor-pink?style=for-the-badge&logo=github-sponsors&logoColor=white)](https://github.com/sponsors/unclecode)
+[![Current Sponsors](https://img.shields.io/github/sponsors/unclecode?style=for-the-badge&logo=github&label=Current%20Sponsors&color=green)](https://github.com/sponsors/unclecode)
+
+</div>
+
+### 🤝 Sponsorship Tiers
+
+- **🌱 Believer ($5/mo)**: Join the movement for data democratization
+- **🚀 Builder ($50/mo)**: Get priority support and early feature access  
+- **💼 Growing Team ($500/mo)**: Bi-weekly syncs and optimization help
+- **🏢 Data Infrastructure Partner ($2000/mo)**: Full partnership with dedicated support
+
+**Why sponsor?** Every tier includes real benefits. No more rate-limited APIs. Own your data pipeline. Build data sovereignty together.
+
+[View All Tiers & Benefits →](https://github.com/sponsors/unclecode)
+
+### 🏆 Our Sponsors
+
+#### 👑 Founding Sponsors (First 50)
+*Be part of history - [Become a Founding Sponsor](https://github.com/sponsors/unclecode)*
+
+<!-- Founding sponsors will be permanently recognized here -->
+
+#### Current Sponsors
+Thank you to all our sponsors who make this project possible!
+
+<!-- Sponsors will be automatically added here -->
+
+## 🗾 Mission
+
+Our mission is to unlock the value of personal and enterprise data by transforming digital footprints into structured, tradeable assets. Crawl4AI empowers individuals and organizations with open-source tools to extract and structure data, fostering a shared data economy.  
+
+We envision a future where AI is powered by real human knowledge, ensuring data creators directly benefit from their contributions. By democratizing data and enabling ethical sharing, we are laying the foundation for authentic AI advancement.
+
+<details>
+<summary>🔑 <strong>Key Opportunities</strong></summary>
+ 
+- **Data Capitalization**: Transform digital footprints into measurable, valuable assets.  
+- **Authentic AI Data**: Provide AI systems with real human insights.  
+- **Shared Economy**: Create a fair data marketplace that benefits data creators.  
+
+</details>
+
+<details>
+<summary>🚀 <strong>Development Pathway</strong></summary>
+
+1. **Open-Source Tools**: Community-driven platforms for transparent data extraction.  
+2. **Digital Asset Structuring**: Tools to organize and value digital knowledge.  
+3. **Ethical Data Marketplace**: A secure, fair platform for exchanging structured data.  
+
+For more details, see our [full mission statement](./MISSION.md).
+</details>
+
+## Star History
+
+[![Star History Chart](https://api.star-history.com/svg?repos=unclecode/crawl4ai&type=Date)](https://star-history.com/#unclecode/crawl4ai&Date)

README.md

@@ -10,6 +10,7 @@
 [![PyPI version](https://badge.fury.io/py/crawl4ai.svg)](https://badge.fury.io/py/crawl4ai)
 [![Python Version](https://img.shields.io/pypi/pyversions/crawl4ai)](https://pypi.org/project/crawl4ai/)
 [![Downloads](https://static.pepy.tech/badge/crawl4ai/month)](https://pepy.tech/project/crawl4ai)
+[![GitHub Sponsors](https://img.shields.io/github/sponsors/unclecode?style=flat&logo=GitHub-Sponsors&label=Sponsors&color=pink)](https://github.com/sponsors/unclecode)
 
 <p align="center">
     <a href="https://x.com/crawl4ai">
@@ -24,32 +25,33 @@
   </p>
 </div>
 
-Crawl4AI is the #1 trending GitHub repository, actively maintained by a vibrant community. It delivers blazing-fast, AI-ready web crawling tailored for LLMs, AI agents, and data pipelines. Open source, flexible, and built for real-time performance, Crawl4AI empowers developers with unmatched speed, precision, and deployment ease.  
+Crawl4AI turns the web into clean, LLM ready Markdown for RAG, agents, and data pipelines. Fast, controllable, battle tested by a 50k+ star community.
 
-[✨ Check out latest update v0.7.0](#-recent-updates)
+[✨ Check out latest update v0.7.3](#-recent-updates)
 
-🎉 **Version 0.7.0 is now available!** The Adaptive Intelligence Update introduces groundbreaking features: Adaptive Crawling that learns website patterns, Virtual Scroll support for infinite pages, intelligent Link Preview with 3-layer scoring, Async URL Seeder for massive discovery, and significant performance improvements. [Read the release notes →](https://docs.crawl4ai.com/blog/release-v0.7.0)
+✨ New in v0.7.3: Undetected Browser Support, Multi-URL Configurations, Memory Monitoring, Enhanced Table Extraction, GitHub Sponsors. [Release notes →](https://github.com/unclecode/crawl4ai/blob/main/docs/blog/release-v0.7.3.md)
 
 <details>
-<summary>🤓 <strong>My Personal Story</strong></summary>
+  <summary>🤓 <strong>My Personal Story</strong></summary>
 
-My journey with computers started in childhood when my dad, a computer scientist, introduced me to an Amstrad computer. Those early days sparked a fascination with technology, leading me to pursue computer science and specialize in NLP during my postgraduate studies. It was during this time that I first delved into web crawling, building tools to help researchers organize papers and extract information from publications a challenging yet rewarding experience that honed my skills in data extraction.
+I grew up on an Amstrad, thanks to my dad, and never stopped building. In grad school I specialized in NLP and built crawlers for research. That’s where I learned how much extraction matters.
 
-Fast forward to 2023, I was working on a tool for a project and needed a crawler to convert a webpage into markdown. While exploring solutions, I found one that claimed to be open-source but required creating an account and generating an API token. Worse, it turned out to be a SaaS model charging $16, and its quality didn’t meet my standards. Frustrated, I realized this was a deeper problem. That frustration turned into turbo anger mode, and I decided to build my own solution. In just a few days, I created Crawl4AI. To my surprise, it went viral, earning thousands of GitHub stars and resonating with a global community.
+In 2023, I needed web-to-Markdown. The “open source” option wanted an account, API token, and $16, and still under-delivered. I went turbo anger mode, built Crawl4AI in days, and it went viral. Now it’s the most-starred crawler on GitHub.
 
-I made Crawl4AI open-source for two reasons. First, it’s my way of giving back to the open-source community that has supported me throughout my career. Second, I believe data should be accessible to everyone, not locked behind paywalls or monopolized by a few. Open access to data lays the foundation for the democratization of AI, a vision where individuals can train their own models and take ownership of their information. This library is the first step in a larger journey to create the best open-source data extraction and generation tool the world has ever seen, built collaboratively by a passionate community.
-
-Thank you to everyone who has supported this project, used it, and shared feedback. Your encouragement motivates me to dream even bigger. Join us, file issues, submit PRs, or spread the word. Together, we can build a tool that truly empowers people to access their own data and reshape the future of AI.
+I made it open source for **availability**, anyone can use it without a gate. Now I’m building the platform for **affordability**, anyone can run serious crawls without breaking the bank. If that resonates, join in, send feedback, or just crawl something amazing.
 </details>
 
-## 🧐 Why Crawl4AI?
 
-1. **Built for LLMs**: Creates smart, concise Markdown optimized for RAG and fine-tuning applications.  
-2. **Lightning Fast**: Delivers results 6x faster with real-time, cost-efficient performance.  
-3. **Flexible Browser Control**: Offers session management, proxies, and custom hooks for seamless data access.  
-4. **Heuristic Intelligence**: Uses advanced algorithms for efficient extraction, reducing reliance on costly models.  
-5. **Open Source & Deployable**: Fully open-source with no API keys—ready for Docker and cloud integration.  
-6. **Thriving Community**: Actively maintained by a vibrant community and the #1 trending GitHub repository.
+<details>
+  <summary>Why developers pick Crawl4AI</summary>
+
+- **LLM ready output**, smart Markdown with headings, tables, code, citation hints
+- **Fast in practice**, async browser pool, caching, minimal hops
+- **Full control**, sessions, proxies, cookies, user scripts, hooks
+- **Adaptive intelligence**, learns site patterns, explores only what matters
+- **Deploy anywhere**, zero keys, CLI and Docker, cloud friendly
+</details>
+
 
 ## 🚀 Quick Start 
 
@@ -101,6 +103,33 @@ crwl https://docs.crawl4ai.com --deep-crawl bfs --max-pages 10
 crwl https://www.example.com/products -q "Extract all product prices"
 ```
 
+## 💖 Support Crawl4AI
+
+> 🎉 **Sponsorship Program Now Open!** After powering 51K+ developers and 1 year of growth, Crawl4AI is launching dedicated support for **startups** and **enterprises**. Be among the first 50 **Founding Sponsors** for permanent recognition in our Hall of Fame.
+
+Crawl4AI is the #1 trending open-source web crawler on GitHub. Your support keeps it independent, innovative, and free for the community — while giving you direct access to premium benefits.
+
+<div align="">
+  
+[![Become a Sponsor](https://img.shields.io/badge/Become%20a%20Sponsor-pink?style=for-the-badge&logo=github-sponsors&logoColor=white)](https://github.com/sponsors/unclecode)  
+[![Current Sponsors](https://img.shields.io/github/sponsors/unclecode?style=for-the-badge&logo=github&label=Current%20Sponsors&color=green)](https://github.com/sponsors/unclecode)
+
+</div>
+
+### 🤝 Sponsorship Tiers
+
+- **🌱 Believer ($5/mo)** — Join the movement for data democratization  
+- **🚀 Builder ($50/mo)** — Priority support & early access to features  
+- **💼 Growing Team ($500/mo)** — Bi-weekly syncs & optimization help  
+- **🏢 Data Infrastructure Partner ($2000/mo)** — Full partnership with dedicated support  
+  *Custom arrangements available - see [SPONSORS.md](SPONSORS.md) for details & contact*
+
+**Why sponsor?**  
+No rate-limited APIs. No lock-in. Build and own your data pipeline with direct guidance from the creator of Crawl4AI.
+
+[See All Tiers & Benefits →](https://github.com/sponsors/unclecode)
+
+
 ## ✨ Features 
 
 <details>
@@ -280,12 +309,6 @@ docker run -d -p 11235:11235 --name crawl4ai --shm-size=1g unclecode/crawl4ai:0.
 # Visit the playground at http://localhost:11235/playground
 ```
 
-For complete documentation, see our [Docker Deployment Guide](https://docs.crawl4ai.com/core/docker-deployment/).
-
-</details>
-
----
-
 ### Quick Test
 
 Run a quick test (works for both Docker options):
@@ -316,10 +339,11 @@ For more examples, see our [Docker Examples](https://github.com/unclecode/crawl4
 
 </details>
 
+---
 
 ## 🔬 Advanced Usage Examples 🔬
 
-You can check the project structure in the directory [https://github.com/unclecode/crawl4ai/docs/examples](docs/examples). Over there, you can find a variety of examples; here, some popular examples are shared.
+You can check the project structure in the directory [docs/examples](https://github.com/unclecode/crawl4ai/tree/main/docs/examples). Over there, you can find a variety of examples; here, some popular examples are shared.
 
 <details>
 <summary>📝 <strong>Heuristic Markdown Generation with Clean and Fit Markdown</strong></summary>
@@ -478,7 +502,7 @@ if __name__ == "__main__":
 </details>
 
 <details>
-<summary>🤖 <strong>Using You own Browser with Custom User Profile</strong></summary>
+<summary>🤖 <strong>Using Your own Browser with Custom User Profile</strong></summary>
 
 ```python
 import os, sys
@@ -518,7 +542,89 @@ async def test_news_crawl():
 
 ## ✨ Recent Updates
 
-### Version 0.7.0 Release Highlights - The Adaptive Intelligence Update
+<details>
+<summary><strong>Version 0.7.3 Release Highlights - The Multi-Config Intelligence Update</strong></summary>
+
+- **🕵️ Undetected Browser Support**: Bypass sophisticated bot detection systems:
+  ```python
+  from crawl4ai import AsyncWebCrawler, BrowserConfig
+  
+  browser_config = BrowserConfig(
+      browser_type="undetected",  # Use undetected Chrome
+      headless=True,              # Can run headless with stealth
+      extra_args=[
+          "--disable-blink-features=AutomationControlled",
+          "--disable-web-security"
+      ]
+  )
+  
+  async with AsyncWebCrawler(config=browser_config) as crawler:
+      result = await crawler.arun("https://protected-site.com")
+  # Successfully bypass Cloudflare, Akamai, and custom bot detection
+  ```
+
+- **🎨 Multi-URL Configuration**: Different strategies for different URL patterns in one batch:
+  ```python
+  from crawl4ai import CrawlerRunConfig, MatchMode
+  
+  configs = [
+      # Documentation sites - aggressive caching
+      CrawlerRunConfig(
+          url_matcher=["*docs*", "*documentation*"],
+          cache_mode="write",
+          markdown_generator_options={"include_links": True}
+      ),
+      
+      # News/blog sites - fresh content
+      CrawlerRunConfig(
+          url_matcher=lambda url: 'blog' in url or 'news' in url,
+          cache_mode="bypass"
+      ),
+      
+      # Fallback for everything else
+      CrawlerRunConfig()
+  ]
+  
+  results = await crawler.arun_many(urls, config=configs)
+  # Each URL gets the perfect configuration automatically
+  ```
+
+- **🧠 Memory Monitoring**: Track and optimize memory usage during crawling:
+  ```python
+  from crawl4ai.memory_utils import MemoryMonitor
+  
+  monitor = MemoryMonitor()
+  monitor.start_monitoring()
+  
+  results = await crawler.arun_many(large_url_list)
+  
+  report = monitor.get_report()
+  print(f"Peak memory: {report['peak_mb']:.1f} MB")
+  print(f"Efficiency: {report['efficiency']:.1f}%")
+  # Get optimization recommendations
+  ```
+
+- **📊 Enhanced Table Extraction**: Direct DataFrame conversion from web tables:
+  ```python
+  result = await crawler.arun("https://site-with-tables.com")
+  
+  # New way - direct table access
+  if result.tables:
+      import pandas as pd
+      for table in result.tables:
+          df = pd.DataFrame(table['data'])
+          print(f"Table: {df.shape[0]} rows × {df.shape[1]} columns")
+  ```
+
+- **💰 GitHub Sponsors**: 4-tier sponsorship system for project sustainability
+- **🐳 Docker LLM Flexibility**: Configure providers via environment variables
+
+[Full v0.7.3 Release Notes →](https://github.com/unclecode/crawl4ai/blob/main/docs/blog/release-v0.7.3.md)
+
+</details>
+
+<details>
+<summary><strong>Version 0.7.0 Release Highlights - The Adaptive Intelligence Update</strong></summary>
 
 - **🧠 Adaptive Crawling**: Your crawler now learns and adapts to website patterns automatically:
   ```python
@@ -583,97 +689,14 @@ async def test_news_crawl():
 
 Read the full details in our [0.7.0 Release Notes](https://docs.crawl4ai.com/blog/release-v0.7.0) or check the [CHANGELOG](https://github.com/unclecode/crawl4ai/blob/main/CHANGELOG.md).
 
-### Previous Version: 0.6.0 Release Highlights
-
-- **🌎 World-aware Crawling**: Set geolocation, language, and timezone for authentic locale-specific content:
-  ```python
-    crun_cfg = CrawlerRunConfig(
-        url="https://browserleaks.com/geo",          # test page that shows your location
-        locale="en-US",                              # Accept-Language & UI locale
-        timezone_id="America/Los_Angeles",           # JS Date()/Intl timezone
-        geolocation=GeolocationConfig(                 # override GPS coords
-            latitude=34.0522,
-            longitude=-118.2437,
-            accuracy=10.0,
-        )
-    )
-  ```
-
-- **📊 Table-to-DataFrame Extraction**: Extract HTML tables directly to CSV or pandas DataFrames:
-  ```python
-    crawler = AsyncWebCrawler(config=browser_config)
-    await crawler.start()
-
-    try:
-        # Set up scraping parameters
-        crawl_config = CrawlerRunConfig(
-            table_score_threshold=8,  # Strict table detection
-        )
-
-        # Execute market data extraction
-        results: List[CrawlResult] = await crawler.arun(
-            url="https://coinmarketcap.com/?page=1", config=crawl_config
-        )
-
-        # Process results
-        raw_df = pd.DataFrame()
-        for result in results:
-            if result.success and result.media["tables"]:
-                raw_df = pd.DataFrame(
-                    result.media["tables"][0]["rows"],
-                    columns=result.media["tables"][0]["headers"],
-                )
-                break
-        print(raw_df.head())
-
-    finally:
-        await crawler.stop()
-  ```
-
-- **🚀 Browser Pooling**: Pages launch hot with pre-warmed browser instances for lower latency and memory usage
-
-- **🕸️ Network and Console Capture**: Full traffic logs and MHTML snapshots for debugging:
-  ```python
-  crawler_config = CrawlerRunConfig(
-      capture_network=True,
-      capture_console=True,
-      mhtml=True
-  )
-  ```
-
-- **🔌 MCP Integration**: Connect to AI tools like Claude Code through the Model Context Protocol
-  ```bash
-  # Add Crawl4AI to Claude Code
-  claude mcp add --transport sse c4ai-sse http://localhost:11235/mcp/sse
-  ```
-
-- **🖥️ Interactive Playground**: Test configurations and generate API requests with the built-in web interface at `http://localhost:11235//playground`
-
-- **🐳 Revamped Docker Deployment**: Streamlined multi-architecture Docker image with improved resource efficiency
-
-- **📱 Multi-stage Build System**: Optimized Dockerfile with platform-specific performance enhancements
-
-
-### Previous Version: 0.5.0 Major Release Highlights
-
--   **🚀 Deep Crawling System**: Explore websites beyond initial URLs with BFS, DFS, and BestFirst strategies
--   **⚡ Memory-Adaptive Dispatcher**: Dynamically adjusts concurrency based on system memory
--   **🔄 Multiple Crawling Strategies**: Browser-based and lightweight HTTP-only crawlers
--   **💻 Command-Line Interface**: New `crwl` CLI provides convenient terminal access
--   **👤 Browser Profiler**: Create and manage persistent browser profiles
--   **🧠 Crawl4AI Coding Assistant**: AI-powered coding assistant
--   **🏎️ LXML Scraping Mode**: Fast HTML parsing using the `lxml` library
--   **🌐 Proxy Rotation**: Built-in support for proxy switching
--   **🤖 LLM Content Filter**: Intelligent markdown generation using LLMs
--   **📄 PDF Processing**: Extract text, images, and metadata from PDF files
-
-Read the full details in our [0.5.0 Release Notes](https://docs.crawl4ai.com/blog/releases/0.5.0.html).
+</details>
 
 ## Version Numbering in Crawl4AI
 
 Crawl4AI follows standard Python version numbering conventions (PEP 440) to help users understand the stability and features of each release.
 
-### Version Numbers Explained
+<details>
+<summary>📈 <strong>Version Numbers Explained</strong></summary>
 
 Our version numbers follow this pattern: `MAJOR.MINOR.PATCH` (e.g., 0.4.3)
 
@@ -710,6 +733,8 @@ We use pre-releases to:
 
 For production environments, we recommend using the stable version. For testing new features, you can opt-in to pre-releases using the `--pre` flag.
 
+</details>
+
 ## 📖 Documentation & Roadmap 
 
 > 🚨 **Documentation Update Alert**: We're undertaking a major documentation overhaul next week to reflect recent updates and improvements. Stay tuned for a more comprehensive and up-to-date guide!
@@ -722,16 +747,16 @@ To check our development plans and upcoming features, visit our [Roadmap](https:
 <summary>📈 <strong>Development TODOs</strong></summary>
 
 - [x] 0. Graph Crawler: Smart website traversal using graph search algorithms for comprehensive nested page extraction
-- [ ] 1. Question-Based Crawler: Natural language driven web discovery and content extraction
-- [ ] 2. Knowledge-Optimal Crawler: Smart crawling that maximizes knowledge while minimizing data extraction
-- [ ] 3. Agentic Crawler: Autonomous system for complex multi-step crawling operations
-- [ ] 4. Automated Schema Generator: Convert natural language to extraction schemas
-- [ ] 5. Domain-Specific Scrapers: Pre-configured extractors for common platforms (academic, e-commerce)
-- [ ] 6. Web Embedding Index: Semantic search infrastructure for crawled content
-- [ ] 7. Interactive Playground: Web UI for testing, comparing strategies with AI assistance
-- [ ] 8. Performance Monitor: Real-time insights into crawler operations
+- [x] 1. Question-Based Crawler: Natural language driven web discovery and content extraction
+- [x] 2. Knowledge-Optimal Crawler: Smart crawling that maximizes knowledge while minimizing data extraction
+- [x] 3. Agentic Crawler: Autonomous system for complex multi-step crawling operations
+- [x] 4. Automated Schema Generator: Convert natural language to extraction schemas
+- [x] 5. Domain-Specific Scrapers: Pre-configured extractors for common platforms (academic, e-commerce)
+- [x] 6. Web Embedding Index: Semantic search infrastructure for crawled content
+- [x] 7. Interactive Playground: Web UI for testing, comparing strategies with AI assistance
+- [x] 8. Performance Monitor: Real-time insights into crawler operations
 - [ ] 9. Cloud Integration: One-click deployment solutions across cloud providers
-- [ ] 10. Sponsorship Program: Structured support system with tiered benefits
+- [x] 10. Sponsorship Program: Structured support system with tiered benefits
 - [ ] 11. Educational Content: "How to Crawl" video series and interactive tutorials
 
 </details>
@@ -746,12 +771,13 @@ Here's the updated license section:
 
 ## 📄 License & Attribution
 
-This project is licensed under the Apache License 2.0 with a required attribution clause. See the [Apache 2.0 License](https://github.com/unclecode/crawl4ai/blob/main/LICENSE) file for details.
+This project is licensed under the Apache License 2.0, attribution is recommended via the badges below. See the [Apache 2.0 License](https://github.com/unclecode/crawl4ai/blob/main/LICENSE) file for details.
 
 ### Attribution Requirements
 When using Crawl4AI, you must include one of the following attribution methods:
 
-#### 1. Badge Attribution (Recommended)
+<details>
+<summary>📈 <strong>1. Badge Attribution (Recommended)</strong></summary>
 Add one of these badges to your README, documentation, or website:
 
 | Theme | Badge |
@@ -790,11 +816,15 @@ HTML code for adding the badges:
 </a>
 ```
 
-#### 2. Text Attribution
+</details>
+
+<details>
+<summary>📖 <strong>2. Text Attribution</strong></summary>
 Add this line to your documentation:
 ```
 This project uses Crawl4AI (https://github.com/unclecode/crawl4ai) for web data extraction.
 ```
+</details>
 
 ## 📚 Citation
 

SPONSORS.md

@@ -0,0 +1,65 @@
+# 💖 Sponsors & Supporters
+
+Thank you to everyone supporting Crawl4AI! Your sponsorship helps keep this project open-source and actively maintained.
+
+## 👑 Founding Sponsors
+*The first 50 sponsors who believed in our vision - permanently recognized*
+
+<!-- Founding sponsors will be listed here with special recognition -->
+🎉 **Become a Founding Sponsor!** Only [X/50] spots remaining! [Join now →](https://github.com/sponsors/unclecode)
+
+---
+
+## 🏢 Data Infrastructure Partners ($2000/month)
+*These organizations are building their data sovereignty with Crawl4AI at the core*
+
+<!-- Data Infrastructure Partners will be listed here -->
+*Be the first Data Infrastructure Partner! [Join us →](https://github.com/sponsors/unclecode)*
+
+---
+
+## 💼 Growing Teams ($500/month)
+*Teams scaling their data extraction with Crawl4AI*
+
+<!-- Growing Teams will be listed here -->
+*Your team could be here! [Become a sponsor →](https://github.com/sponsors/unclecode)*
+
+---
+
+## 🚀 Builders ($50/month)
+*Developers and entrepreneurs building with Crawl4AI*
+
+<!-- Builders will be listed here -->
+*Join the builders! [Start sponsoring →](https://github.com/sponsors/unclecode)*
+
+---
+
+## 🌱 Believers ($5/month)
+*The community supporting data democratization*
+
+<!-- Believers will be listed here -->
+*Thank you to all our community believers!*
+
+---
+
+## 🤝 Want to Sponsor?
+
+Crawl4AI is the #1 trending open-source web crawler. We're building the future of data extraction - where organizations own their data pipelines instead of relying on rate-limited APIs.
+
+### Available Sponsorship Tiers:
+- **🌱 Believer** ($5/mo) - Support the movement
+- **🚀 Builder** ($50/mo) - Priority support & early access
+- **💼 Growing Team** ($500/mo) - Bi-weekly syncs & optimization
+- **🏢 Data Infrastructure Partner** ($2000/mo) - Full partnership & dedicated support
+
+[View all tiers and benefits →](https://github.com/sponsors/unclecode)
+
+### Enterprise & Custom Partnerships
+
+Building data extraction at scale? Need dedicated support or infrastructure? Let's talk about a custom partnership.
+
+📧 Contact: [hello@crawl4ai.com](mailto:hello@crawl4ai.com) | 📅 [Schedule a call](https://calendar.app.google/rEpvi2UBgUQjWHfJ9)
+
+---
+
+*This list is updated regularly. Sponsors at $50+ tiers can submit their logos via [hello@crawl4ai.com](mailto:hello@crawl4ai.com)*

crawl4ai/__init__.py

@@ -3,12 +3,12 @@ import warnings
 
 from .async_webcrawler import AsyncWebCrawler, CacheMode
 # MODIFIED: Add SeedingConfig and VirtualScrollConfig here
-from .async_configs import BrowserConfig, CrawlerRunConfig, HTTPCrawlerConfig, LLMConfig, ProxyConfig, GeolocationConfig, SeedingConfig, VirtualScrollConfig, LinkPreviewConfig
+from .async_configs import BrowserConfig, CrawlerRunConfig, HTTPCrawlerConfig, LLMConfig, ProxyConfig, GeolocationConfig, SeedingConfig, VirtualScrollConfig, LinkPreviewConfig, MatchMode
 
 from .content_scraping_strategy import (
     ContentScrapingStrategy,
-    WebScrapingStrategy,
     LXMLWebScrapingStrategy,
+    WebScrapingStrategy,  # Backward compatibility alias
 )
 from .async_logger import (
     AsyncLoggerBase,
@@ -88,6 +88,13 @@ from .script import (
     ErrorDetail
 )
 
+# Browser Adapters
+from .browser_adapter import (
+    BrowserAdapter,
+    PlaywrightAdapter,
+    UndetectedAdapter
+)
+
 from .utils import (
     start_colab_display_server,
     setup_colab_environment
@@ -132,6 +139,7 @@ __all__ = [
     "CrawlResult",
     "CrawlerHub",
     "CacheMode",
+    "MatchMode",
     "ContentScrapingStrategy",
     "WebScrapingStrategy",
     "LXMLWebScrapingStrategy",
@@ -173,6 +181,10 @@ __all__ = [
     "CompilationResult",
     "ValidationResult",
     "ErrorDetail",
+    # Browser Adapters
+    "BrowserAdapter",
+    "PlaywrightAdapter", 
+    "UndetectedAdapter",
     "LinkPreviewConfig"
 ]
 

crawl4ai/__version__.py

@@ -1,7 +1,7 @@
 # crawl4ai/__version__.py
 
 # This is the version that will be used for stable releases
-__version__ = "0.7.1"
+__version__ = "0.7.3"
 
 # For nightly builds, this gets set during build process
 __nightly_version__ = None

crawl4ai/async_configs.py

@@ -18,17 +18,24 @@ from .extraction_strategy import ExtractionStrategy, LLMExtractionStrategy
 from .chunking_strategy import ChunkingStrategy, RegexChunking
 
 from .markdown_generation_strategy import MarkdownGenerationStrategy, DefaultMarkdownGenerator
-from .content_scraping_strategy import ContentScrapingStrategy, WebScrapingStrategy, LXMLWebScrapingStrategy
+from .content_scraping_strategy import ContentScrapingStrategy, LXMLWebScrapingStrategy
 from .deep_crawling import DeepCrawlStrategy
 
 from .cache_context import CacheMode
 from .proxy_strategy import ProxyRotationStrategy
 
-from typing import Union, List
+from typing import Union, List, Callable
 import inspect
 from typing import Any, Dict, Optional
 from enum import Enum
 
+# Type alias for URL matching
+UrlMatcher = Union[str, Callable[[str], bool], List[Union[str, Callable[[str], bool]]]]
+
+class MatchMode(Enum):
+    OR = "or"
+    AND = "and"
+
 # from .proxy_strategy import ProxyConfig
 
 
@@ -383,6 +390,8 @@ class BrowserConfig:
         light_mode (bool): Disables certain background features for performance gains. Default: False.
         extra_args (list): Additional command-line arguments passed to the browser.
                            Default: [].
+        enable_stealth (bool): If True, applies playwright-stealth to bypass basic bot detection.
+                              Cannot be used with use_undetected browser mode. Default: False.
     """
 
     def __init__(
@@ -423,6 +432,7 @@ class BrowserConfig:
         extra_args: list = None,
         debugging_port: int = 9222,
         host: str = "localhost",
+        enable_stealth: bool = False,
     ):
         self.browser_type = browser_type
         self.headless = headless 
@@ -438,6 +448,10 @@ class BrowserConfig:
             self.chrome_channel = ""
         self.proxy = proxy
         self.proxy_config = proxy_config
+        if isinstance(self.proxy_config, dict):
+            self.proxy_config = ProxyConfig.from_dict(self.proxy_config)
+        if isinstance(self.proxy_config, str):
+            self.proxy_config = ProxyConfig.from_string(self.proxy_config)
 
 
         self.viewport_width = viewport_width
@@ -463,6 +477,7 @@ class BrowserConfig:
         self.verbose = verbose
         self.debugging_port = debugging_port
         self.host = host
+        self.enable_stealth = enable_stealth
 
         fa_user_agenr_generator = ValidUAGenerator()
         if self.user_agent_mode == "random":
@@ -494,6 +509,13 @@ class BrowserConfig:
         # If persistent context is requested, ensure managed browser is enabled
         if self.use_persistent_context:
             self.use_managed_browser = True
+            
+        # Validate stealth configuration
+        if self.enable_stealth and self.use_managed_browser and self.browser_mode == "builtin":
+            raise ValueError(
+                "enable_stealth cannot be used with browser_mode='builtin'. "
+                "Stealth mode requires a dedicated browser instance."
+            )
 
     @staticmethod
     def from_kwargs(kwargs: dict) -> "BrowserConfig":
@@ -530,6 +552,7 @@ class BrowserConfig:
             extra_args=kwargs.get("extra_args", []),
             debugging_port=kwargs.get("debugging_port", 9222),
             host=kwargs.get("host", "localhost"),
+            enable_stealth=kwargs.get("enable_stealth", False),
         )
 
     def to_dict(self):
@@ -564,6 +587,7 @@ class BrowserConfig:
             "verbose": self.verbose,
             "debugging_port": self.debugging_port,
             "host": self.host,
+            "enable_stealth": self.enable_stealth,
         }
 
                 
@@ -862,7 +886,7 @@ class CrawlerRunConfig():
         parser_type (str): Type of parser to use for HTML parsing.
                            Default: "lxml".
         scraping_strategy (ContentScrapingStrategy): Scraping strategy to use.
-                           Default: WebScrapingStrategy.
+                           Default: LXMLWebScrapingStrategy.
         proxy_config (ProxyConfig or dict or None): Detailed proxy configuration, e.g. {"server": "...", "username": "..."}.
                                      If None, no additional proxy config. Default: None.
 
@@ -1113,6 +1137,9 @@ class CrawlerRunConfig():
         link_preview_config: Union[LinkPreviewConfig, Dict[str, Any]] = None,
         # Virtual Scroll Parameters
         virtual_scroll_config: Union[VirtualScrollConfig, Dict[str, Any]] = None,
+        # URL Matching Parameters
+        url_matcher: Optional[UrlMatcher] = None,
+        match_mode: MatchMode = MatchMode.OR,
         # Experimental Parameters
         experimental: Dict[str, Any] = None,
     ):
@@ -1136,6 +1163,11 @@ class CrawlerRunConfig():
         self.parser_type = parser_type
         self.scraping_strategy = scraping_strategy or LXMLWebScrapingStrategy()
         self.proxy_config = proxy_config
+        if isinstance(proxy_config, dict):
+            self.proxy_config = ProxyConfig.from_dict(proxy_config)
+        if isinstance(proxy_config, str):
+            self.proxy_config = ProxyConfig.from_string(proxy_config)
+
         self.proxy_rotation_strategy = proxy_rotation_strategy
         
         # Browser Location and Identity Parameters
@@ -1266,6 +1298,10 @@ class CrawlerRunConfig():
         else:
             raise ValueError("virtual_scroll_config must be VirtualScrollConfig object or dict")
         
+        # URL Matching Parameters
+        self.url_matcher = url_matcher
+        self.match_mode = match_mode
+        
         # Experimental Parameters
         self.experimental = experimental or {}
         
@@ -1321,6 +1357,51 @@ class CrawlerRunConfig():
             if "compilation error" not in str(e).lower():
                 raise ValueError(f"Failed to compile C4A script: {str(e)}")
             raise
+    
+    def is_match(self, url: str) -> bool:
+        """Check if this config matches the given URL.
+        
+        Args:
+            url: The URL to check against this config's matcher
+            
+        Returns:
+            bool: True if this config should be used for the URL or if no matcher is set.
+        """
+        if self.url_matcher is None:
+            return True
+            
+        if callable(self.url_matcher):
+            # Single function matcher
+            return self.url_matcher(url)
+        
+        elif isinstance(self.url_matcher, str):
+            # Single pattern string
+            from fnmatch import fnmatch
+            return fnmatch(url, self.url_matcher)
+        
+        elif isinstance(self.url_matcher, list):
+            # List of mixed matchers
+            if not self.url_matcher:  # Empty list
+                return False
+                
+            results = []
+            for matcher in self.url_matcher:
+                if callable(matcher):
+                    results.append(matcher(url))
+                elif isinstance(matcher, str):
+                    from fnmatch import fnmatch
+                    results.append(fnmatch(url, matcher))
+                else:
+                    # Skip invalid matchers
+                    continue
+            
+            # Apply match mode logic
+            if self.match_mode == MatchMode.OR:
+                return any(results) if results else False
+            else:  # AND mode
+                return all(results) if results else False
+        
+        return False
 
 
     def __getattr__(self, name):
@@ -1443,6 +1524,9 @@ class CrawlerRunConfig():
             # Link Extraction Parameters
             link_preview_config=kwargs.get("link_preview_config"),
             url=kwargs.get("url"),
+            # URL Matching Parameters
+            url_matcher=kwargs.get("url_matcher"),
+            match_mode=kwargs.get("match_mode", MatchMode.OR),
             # Experimental Parameters 
             experimental=kwargs.get("experimental"),
         )
@@ -1540,6 +1624,8 @@ class CrawlerRunConfig():
             "deep_crawl_strategy": self.deep_crawl_strategy,
             "link_preview_config": self.link_preview_config.to_dict() if self.link_preview_config else None,
             "url": self.url,
+            "url_matcher": self.url_matcher,
+            "match_mode": self.match_mode,
             "experimental": self.experimental,
         }
 

crawl4ai/async_crawler_strategy.py

@@ -21,6 +21,7 @@ from .async_logger import AsyncLogger
 from .ssl_certificate import SSLCertificate
 from .user_agent_generator import ValidUAGenerator
 from .browser_manager import BrowserManager
+from .browser_adapter import BrowserAdapter, PlaywrightAdapter, UndetectedAdapter
 
 import aiofiles
 import aiohttp
@@ -71,7 +72,7 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
     """
 
     def __init__(
-        self, browser_config: BrowserConfig = None, logger: AsyncLogger = None, **kwargs
+        self, browser_config: BrowserConfig = None, logger: AsyncLogger = None, browser_adapter: BrowserAdapter = None, **kwargs
     ):
         """
         Initialize the AsyncPlaywrightCrawlerStrategy with a browser configuration.
@@ -80,11 +81,16 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
             browser_config (BrowserConfig): Configuration object containing browser settings.
                                           If None, will be created from kwargs for backwards compatibility.
             logger: Logger instance for recording events and errors.
+            browser_adapter (BrowserAdapter): Browser adapter for handling browser-specific operations.
+                                           If None, defaults to PlaywrightAdapter.
             **kwargs: Additional arguments for backwards compatibility and extending functionality.
         """
         # Initialize browser config, either from provided object or kwargs
         self.browser_config = browser_config or BrowserConfig.from_kwargs(kwargs)
         self.logger = logger
+        
+        # Initialize browser adapter
+        self.adapter = browser_adapter or PlaywrightAdapter()
 
         # Initialize session management
         self._downloaded_files = []
@@ -104,7 +110,9 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
 
         # Initialize browser manager with config
         self.browser_manager = BrowserManager(
-            browser_config=self.browser_config, logger=self.logger
+            browser_config=self.browser_config, 
+            logger=self.logger,
+            use_undetected=isinstance(self.adapter, UndetectedAdapter)
         )
 
     async def __aenter__(self):
@@ -322,7 +330,7 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
         """
 
         try:
-            result = await page.evaluate(wrapper_js)
+            result = await self.adapter.evaluate(page, wrapper_js)
             return result
         except Exception as e:
             if "Error evaluating condition" in str(e):
@@ -367,7 +375,7 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
 
                     # Replace the iframe with a div containing the extracted content
                     _iframe = iframe_content.replace("`", "\\`")
-                    await page.evaluate(
+                    await self.adapter.evaluate(page,
                         f"""
                         () => {{
                             const iframe = document.getElementById('iframe-{i}');
@@ -628,91 +636,16 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
             page.on("requestfailed", handle_request_failed_capture)
 
         # Console Message Capturing
+        handle_console = None
+        handle_error = None
         if config.capture_console_messages:
-            def handle_console_capture(msg):
-                try:
-                    message_type = "unknown"
-                    try:
-                        message_type = msg.type
-                    except:
-                        pass
-                        
-                    message_text = "unknown"
-                    try:
-                        message_text = msg.text
-                    except:
-                        pass
-                        
-                    # Basic console message with minimal content
-                    entry = {
-                        "type": message_type,
-                        "text": message_text,
-                        "timestamp": time.time()
-                    }
-                    
-                    captured_console.append(entry)
-                    
-                except Exception as e:
-                    if self.logger:
-                        self.logger.warning(f"Error capturing console message: {e}", tag="CAPTURE")
-                    # Still add something to the list even on error
-                    captured_console.append({
-                        "type": "console_capture_error", 
-                        "error": str(e), 
-                        "timestamp": time.time()
-                    })
-
-            def handle_pageerror_capture(err):
-                try:
-                    error_message = "Unknown error"
-                    try:
-                        error_message = err.message
-                    except:
-                        pass
-                        
-                    error_stack = ""
-                    try:
-                        error_stack = err.stack
-                    except:
-                        pass
-                        
-                    captured_console.append({
-                        "type": "error",
-                        "text": error_message,
-                        "stack": error_stack,
-                        "timestamp": time.time()
-                    })
-                except Exception as e:
-                    if self.logger:
-                        self.logger.warning(f"Error capturing page error: {e}", tag="CAPTURE")
-                    captured_console.append({
-                        "type": "pageerror_capture_error", 
-                        "error": str(e), 
-                        "timestamp": time.time()
-                    })
-
-            # Add event listeners directly
-            page.on("console", handle_console_capture)
-            page.on("pageerror", handle_pageerror_capture)
+            # Set up console capture using adapter
+            handle_console = await self.adapter.setup_console_capture(page, captured_console)
+            handle_error = await self.adapter.setup_error_capture(page, captured_console)
 
         # Set up console logging if requested
-        if config.log_console:
-            def log_consol(
-                msg, console_log_type="debug"
-            ):  # Corrected the parameter syntax
-                if console_log_type == "error":
-                    self.logger.error(
-                        message=f"Console error: {msg}",  # Use f-string for variable interpolation
-                        tag="CONSOLE"
-                    )
-                elif console_log_type == "debug":
-                    self.logger.debug(
-                        message=f"Console: {msg}",  # Use f-string for variable interpolation
-                        tag="CONSOLE"
-                    )
-
-            page.on("console", log_consol)
-            page.on("pageerror", lambda e: log_consol(e, "error"))
+        # Note: For undetected browsers, console logging won't work directly
+        # but captured messages can still be logged after retrieval
 
         try:
             # Get SSL certificate information if requested and URL is HTTPS
@@ -824,7 +757,7 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
             except Error:
                 visibility_info = await self.check_visibility(page)
 
-                if self.browser_config.config.verbose:
+                if self.browser_config.verbose:
                     self.logger.debug(
                         message="Body visibility info: {info}",
                         tag="DEBUG",
@@ -998,7 +931,7 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
                         await page.wait_for_load_state("domcontentloaded", timeout=5)
                     except PlaywrightTimeoutError:
                         pass
-                    await page.evaluate(update_image_dimensions_js)
+                    await self.adapter.evaluate(page, update_image_dimensions_js)
                 except Exception as e:
                     self.logger.error(
                         message="Error updating image dimensions: {error}",
@@ -1027,7 +960,7 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
                     
                     for selector in selectors:
                         try:
-                            content = await page.evaluate(
+                            content = await self.adapter.evaluate(page,
                                 f"""Array.from(document.querySelectorAll("{selector}"))
                                     .map(el => el.outerHTML)
                                     .join('')"""
@@ -1085,6 +1018,11 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
                 await asyncio.sleep(delay)
                 return await page.content()
 
+            # For undetected browsers, retrieve console messages before returning
+            if config.capture_console_messages and hasattr(self.adapter, 'retrieve_console_messages'):
+                final_messages = await self.adapter.retrieve_console_messages(page)
+                captured_console.extend(final_messages)
+
             # Return complete response
             return AsyncCrawlResponse(
                 html=html,
@@ -1123,8 +1061,13 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
                     page.remove_listener("response", handle_response_capture)
                     page.remove_listener("requestfailed", handle_request_failed_capture)
                 if config.capture_console_messages:
-                    page.remove_listener("console", handle_console_capture)
-                    page.remove_listener("pageerror", handle_pageerror_capture)
+                    # Retrieve any final console messages for undetected browsers
+                    if hasattr(self.adapter, 'retrieve_console_messages'):
+                        final_messages = await self.adapter.retrieve_console_messages(page)
+                        captured_console.extend(final_messages)
+                    
+                    # Clean up console capture
+                    await self.adapter.cleanup_console_capture(page, handle_console, handle_error)
                 
                 # Close the page
                 await page.close()
@@ -1354,7 +1297,7 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
             """
             
             # Execute virtual scroll capture
-            result = await page.evaluate(virtual_scroll_js, config.to_dict())
+            result = await self.adapter.evaluate(page, virtual_scroll_js, config.to_dict())
             
             if result.get("replaced", False):
                 self.logger.success(
@@ -1438,7 +1381,7 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
         remove_overlays_js = load_js_script("remove_overlay_elements")
 
         try:
-            await page.evaluate(
+            await self.adapter.evaluate(page,
                 f"""
                 (() => {{
                     try {{
@@ -1843,7 +1786,7 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
                         # When {script} contains statements (e.g., const link = …; link.click();), 
                         # this forms invalid JavaScript, causing Playwright execution error: SyntaxError: Unexpected token 'const'.
                         # """
-                        result = await page.evaluate(
+                        result = await self.adapter.evaluate(page,
                             f"""
                         (async () => {{
                             try {{
@@ -1965,7 +1908,7 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
             for script in scripts:
                 try:
                     # Execute the script and wait for network idle
-                    result = await page.evaluate(
+                    result = await self.adapter.evaluate(page,
                         f"""
                         (() => {{
                             return new Promise((resolve) => {{
@@ -2049,7 +1992,7 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
         Returns:
             Boolean indicating visibility
         """
-        return await page.evaluate(
+        return await self.adapter.evaluate(page,
             """
             () => {
                 const element = document.body;
@@ -2090,7 +2033,7 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
             Dict containing scroll status and position information
         """
         try:
-            result = await page.evaluate(
+            result = await self.adapter.evaluate(page,
                 f"""() => {{
                     try {{
                         const startX = window.scrollX;
@@ -2147,7 +2090,7 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
         Returns:
             Dict containing width and height of the page
         """
-        return await page.evaluate(
+        return await self.adapter.evaluate(page,
             """
             () => {
                 const {scrollWidth, scrollHeight} = document.documentElement;
@@ -2167,7 +2110,7 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
             bool: True if page needs scrolling
         """
         try:
-            need_scroll = await page.evaluate(
+            need_scroll = await self.adapter.evaluate(page,
                 """
             () => {
                 const scrollHeight = document.documentElement.scrollHeight;
@@ -2186,265 +2129,3 @@ class AsyncPlaywrightCrawlerStrategy(AsyncCrawlerStrategy):
             return True  # Default to scrolling if check fails
 
 
-####################################################################################################
-# HTTP Crawler Strategy
-####################################################################################################
-
-class HTTPCrawlerError(Exception):
-    """Base error class for HTTP crawler specific exceptions"""
-    pass
-
-
-class ConnectionTimeoutError(HTTPCrawlerError):
-    """Raised when connection timeout occurs"""
-    pass
-
-
-class HTTPStatusError(HTTPCrawlerError):
-    """Raised for unexpected status codes"""
-    def __init__(self, status_code: int, message: str):
-        self.status_code = status_code
-        super().__init__(f"HTTP {status_code}: {message}")
-
-
-class AsyncHTTPCrawlerStrategy(AsyncCrawlerStrategy):
-    """
-    Fast, lightweight HTTP-only crawler strategy optimized for memory efficiency.
-    """
-    
-    __slots__ = ('logger', 'max_connections', 'dns_cache_ttl', 'chunk_size', '_session', 'hooks', 'browser_config')
-
-    DEFAULT_TIMEOUT: Final[int] = 30
-    DEFAULT_CHUNK_SIZE: Final[int] = 64 * 1024  
-    DEFAULT_MAX_CONNECTIONS: Final[int] = min(32, (os.cpu_count() or 1) * 4)
-    DEFAULT_DNS_CACHE_TTL: Final[int] = 300
-    VALID_SCHEMES: Final = frozenset({'http', 'https', 'file', 'raw'})
-
-    _BASE_HEADERS: Final = MappingProxyType({
-        'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
-        'Accept-Language': 'en-US,en;q=0.5',
-        'Accept-Encoding': 'gzip, deflate, br',
-        'Connection': 'keep-alive',
-        'Upgrade-Insecure-Requests': '1',
-        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36'
-    })
-    
-    def __init__(
-        self, 
-        browser_config: Optional[HTTPCrawlerConfig] = None,
-        logger: Optional[AsyncLogger] = None,
-        max_connections: int = DEFAULT_MAX_CONNECTIONS,
-        dns_cache_ttl: int = DEFAULT_DNS_CACHE_TTL,
-        chunk_size: int = DEFAULT_CHUNK_SIZE
-    ):
-        """Initialize the HTTP crawler with config"""
-        self.browser_config = browser_config or HTTPCrawlerConfig()
-        self.logger = logger
-        self.max_connections = max_connections
-        self.dns_cache_ttl = dns_cache_ttl
-        self.chunk_size = chunk_size
-        self._session: Optional[aiohttp.ClientSession] = None
-        
-        self.hooks = {
-            k: partial(self._execute_hook, k) 
-            for k in ('before_request', 'after_request', 'on_error')
-        }
-
-        # Set default hooks
-        self.set_hook('before_request', lambda *args, **kwargs: None)
-        self.set_hook('after_request', lambda *args, **kwargs: None)
-        self.set_hook('on_error', lambda *args, **kwargs: None)
-                      
-
-    async def __aenter__(self) -> AsyncHTTPCrawlerStrategy:
-        await self.start()
-        return self
-        
-    async def __aexit__(self, exc_type, exc_val, exc_tb) -> None:
-        await self.close()
-
-    @contextlib.asynccontextmanager
-    async def _session_context(self):
-        try:
-            if not self._session:
-                await self.start()
-            yield self._session
-        finally:
-            pass
-
-    def set_hook(self, hook_type: str, hook_func: Callable) -> None:
-        if hook_type in self.hooks:
-            self.hooks[hook_type] = partial(self._execute_hook, hook_type, hook_func)
-        else:
-            raise ValueError(f"Invalid hook type: {hook_type}")
-
-    async def _execute_hook(
-        self, 
-        hook_type: str, 
-        hook_func: Callable,
-        *args: Any, 
-        **kwargs: Any
-    ) -> Any:
-        if asyncio.iscoroutinefunction(hook_func):
-            return await hook_func(*args, **kwargs)
-        return hook_func(*args, **kwargs)
-
-    async def start(self) -> None:
-        if not self._session:
-            connector = aiohttp.TCPConnector(
-                limit=self.max_connections,
-                ttl_dns_cache=self.dns_cache_ttl,
-                use_dns_cache=True,
-                force_close=False
-            )
-            self._session = aiohttp.ClientSession(
-                headers=dict(self._BASE_HEADERS),
-                connector=connector,
-                timeout=ClientTimeout(total=self.DEFAULT_TIMEOUT)
-            )
-
-    async def close(self) -> None:
-        if self._session and not self._session.closed:
-            try:
-                await asyncio.wait_for(self._session.close(), timeout=5.0)
-            except asyncio.TimeoutError:
-                if self.logger:
-                    self.logger.warning(
-                        message="Session cleanup timed out",
-                        tag="CLEANUP"
-                    )
-            finally:
-                self._session = None
-
-    async def _stream_file(self, path: str) -> AsyncGenerator[memoryview, None]:
-        async with aiofiles.open(path, mode='rb') as f:
-            while chunk := await f.read(self.chunk_size):
-                yield memoryview(chunk)
-
-    async def _handle_file(self, path: str) -> AsyncCrawlResponse:
-        if not os.path.exists(path):
-            raise FileNotFoundError(f"Local file not found: {path}")
-            
-        chunks = []
-        async for chunk in self._stream_file(path):
-            chunks.append(chunk.tobytes().decode('utf-8', errors='replace'))
-            
-        return AsyncCrawlResponse(
-            html=''.join(chunks),
-            response_headers={},
-            status_code=200
-        )
-
-    async def _handle_raw(self, content: str) -> AsyncCrawlResponse:
-        return AsyncCrawlResponse(
-            html=content,
-            response_headers={},
-            status_code=200
-        )
-
-
-    async def _handle_http(
-        self, 
-        url: str, 
-        config: CrawlerRunConfig
-    ) -> AsyncCrawlResponse:
-        async with self._session_context() as session:
-            timeout = ClientTimeout(
-                total=config.page_timeout or self.DEFAULT_TIMEOUT,
-                connect=10,
-                sock_read=30
-            )
-            
-            headers = dict(self._BASE_HEADERS)
-            if self.browser_config.headers:
-                headers.update(self.browser_config.headers)
-
-            request_kwargs = {
-                'timeout': timeout,
-                'allow_redirects': self.browser_config.follow_redirects,
-                'ssl': self.browser_config.verify_ssl,
-                'headers': headers
-            }
-
-            if self.browser_config.method == "POST":
-                if self.browser_config.data:
-                    request_kwargs['data'] = self.browser_config.data
-                if self.browser_config.json:
-                    request_kwargs['json'] = self.browser_config.json
-
-            await self.hooks['before_request'](url, request_kwargs)
-
-            try:
-                async with session.request(self.browser_config.method, url, **request_kwargs) as response:
-                    content = memoryview(await response.read())
-                    
-                    if not (200 <= response.status < 300):
-                        raise HTTPStatusError(
-                            response.status,
-                            f"Unexpected status code for {url}"
-                        )
-                    
-                    encoding = response.charset
-                    if not encoding:
-                        encoding = chardet.detect(content.tobytes())['encoding'] or 'utf-8'                    
-                    
-                    result = AsyncCrawlResponse(
-                        html=content.tobytes().decode(encoding, errors='replace'),
-                        response_headers=dict(response.headers),
-                        status_code=response.status,
-                        redirected_url=str(response.url)
-                    )
-                    
-                    await self.hooks['after_request'](result)
-                    return result
-
-            except aiohttp.ServerTimeoutError as e:
-                await self.hooks['on_error'](e)
-                raise ConnectionTimeoutError(f"Request timed out: {str(e)}")
-                
-            except aiohttp.ClientConnectorError as e:
-                await self.hooks['on_error'](e)
-                raise ConnectionError(f"Connection failed: {str(e)}")
-                
-            except aiohttp.ClientError as e:
-                await self.hooks['on_error'](e)
-                raise HTTPCrawlerError(f"HTTP client error: {str(e)}")
-            
-            except asyncio.exceptions.TimeoutError as e:
-                await self.hooks['on_error'](e)
-                raise ConnectionTimeoutError(f"Request timed out: {str(e)}")
-            
-            except Exception as e:
-                await self.hooks['on_error'](e)
-                raise HTTPCrawlerError(f"HTTP request failed: {str(e)}")
-
-    async def crawl(
-        self, 
-        url: str, 
-        config: Optional[CrawlerRunConfig] = None, 
-        **kwargs
-    ) -> AsyncCrawlResponse:
-        config = config or CrawlerRunConfig.from_kwargs(kwargs)
-        
-        parsed = urlparse(url)
-        scheme = parsed.scheme.rstrip('/')
-        
-        if scheme not in self.VALID_SCHEMES:
-            raise ValueError(f"Unsupported URL scheme: {scheme}")
-            
-        try:
-            if scheme == 'file':
-                return await self._handle_file(parsed.path)
-            elif scheme == 'raw':
-                return await self._handle_raw(parsed.path)
-            else:  # http or https
-                return await self._handle_http(url, config)
-                
-        except Exception as e:
-            if self.logger:
-                self.logger.error(
-                    message="Crawl failed: {error}",
-                    tag="CRAWL",
-                    params={"error": str(e), "url": url}
-                )
-            raise

crawl4ai/async_dispatcher.py

@@ -1,4 +1,4 @@
-from typing import Dict, Optional, List, Tuple
+from typing import Dict, Optional, List, Tuple, Union
 from .async_configs import CrawlerRunConfig
 from .models import (
     CrawlResult,
@@ -22,6 +22,8 @@ from urllib.parse import urlparse
 import random
 from abc import ABC, abstractmethod
 
+from .memory_utils import get_true_memory_usage_percent
+
 
 class RateLimiter:
     def __init__(
@@ -96,11 +98,37 @@ class BaseDispatcher(ABC):
         self.rate_limiter = rate_limiter
         self.monitor = monitor
 
+    def select_config(self, url: str, configs: Union[CrawlerRunConfig, List[CrawlerRunConfig]]) -> Optional[CrawlerRunConfig]:
+        """Select the appropriate config for a given URL.
+        
+        Args:
+            url: The URL to match against
+            configs: Single config or list of configs to choose from
+            
+        Returns:
+            The matching config, or None if no match found
+        """
+        # Single config - return as is
+        if isinstance(configs, CrawlerRunConfig):
+            return configs
+        
+        # Empty list - return None
+        if not configs:
+            return None
+        
+        # Find first matching config
+        for config in configs:
+            if config.is_match(url):
+                return config
+        
+        # No match found - return None to indicate URL should be skipped
+        return None
+
     @abstractmethod
     async def crawl_url(
         self,
         url: str,
-        config: CrawlerRunConfig,
+        config: Union[CrawlerRunConfig, List[CrawlerRunConfig]],
         task_id: str,
         monitor: Optional[CrawlerMonitor] = None,
     ) -> CrawlerTaskResult:
@@ -111,7 +139,7 @@ class BaseDispatcher(ABC):
         self,
         urls: List[str],
         crawler: AsyncWebCrawler,  # noqa: F821
-        config: CrawlerRunConfig,
+        config: Union[CrawlerRunConfig, List[CrawlerRunConfig]],
         monitor: Optional[CrawlerMonitor] = None,
     ) -> List[CrawlerTaskResult]:
         pass
@@ -147,7 +175,7 @@ class MemoryAdaptiveDispatcher(BaseDispatcher):
     async def _memory_monitor_task(self):
         """Background task to continuously monitor memory usage and update state"""
         while True:
-            self.current_memory_percent = psutil.virtual_memory().percent
+            self.current_memory_percent = get_true_memory_usage_percent()
 
             # Enter memory pressure mode if we cross the threshold
             if self.current_memory_percent >= self.memory_threshold_percent:
@@ -200,7 +228,7 @@ class MemoryAdaptiveDispatcher(BaseDispatcher):
     async def crawl_url(
         self,
         url: str,
-        config: CrawlerRunConfig,
+        config: Union[CrawlerRunConfig, List[CrawlerRunConfig]],
         task_id: str,
         retry_count: int = 0,
     ) -> CrawlerTaskResult:
@@ -208,6 +236,37 @@ class MemoryAdaptiveDispatcher(BaseDispatcher):
         error_message = ""
         memory_usage = peak_memory = 0.0
         
+        # Select appropriate config for this URL
+        selected_config = self.select_config(url, config)
+        
+        # If no config matches, return failed result
+        if selected_config is None:
+            error_message = f"No matching configuration found for URL: {url}"
+            if self.monitor:
+                self.monitor.update_task(
+                    task_id, 
+                    status=CrawlStatus.FAILED,
+                    error_message=error_message
+                )
+            
+            return CrawlerTaskResult(
+                task_id=task_id,
+                url=url,
+                result=CrawlResult(
+                    url=url, 
+                    html="", 
+                    metadata={"status": "no_config_match"}, 
+                    success=False, 
+                    error_message=error_message
+                ),
+                memory_usage=0,
+                peak_memory=0,
+                start_time=start_time,
+                end_time=time.time(),
+                error_message=error_message,
+                retry_count=retry_count
+            )
+        
         # Get starting memory for accurate measurement
         process = psutil.Process()
         start_memory = process.memory_info().rss / (1024 * 1024)
@@ -257,8 +316,8 @@ class MemoryAdaptiveDispatcher(BaseDispatcher):
                     retry_count=retry_count + 1
                 )
             
-            # Execute the crawl
-            result = await self.crawler.arun(url, config=config, session_id=task_id)
+            # Execute the crawl with selected config
+            result = await self.crawler.arun(url, config=selected_config, session_id=task_id)
             
             # Measure memory usage
             end_memory = process.memory_info().rss / (1024 * 1024)
@@ -316,7 +375,7 @@ class MemoryAdaptiveDispatcher(BaseDispatcher):
         self,
         urls: List[str],
         crawler: AsyncWebCrawler,
-        config: CrawlerRunConfig,
+        config: Union[CrawlerRunConfig, List[CrawlerRunConfig]],
     ) -> List[CrawlerTaskResult]:
         self.crawler = crawler
         
@@ -348,32 +407,34 @@ class MemoryAdaptiveDispatcher(BaseDispatcher):
                             t.cancel()
                         raise exc
 
-                # If memory pressure is low, start new tasks
-                if not self.memory_pressure_mode and len(active_tasks) < self.max_session_permit:
-                    try:
-                        # Try to get a task with timeout to avoid blocking indefinitely
-                        priority, (url, task_id, retry_count, enqueue_time) = await asyncio.wait_for(
-                            self.task_queue.get(), timeout=0.1
-                        )
-                        
-                        # Create and start the task
-                        task = asyncio.create_task(
-                            self.crawl_url(url, config, task_id, retry_count)
-                        )
-                        active_tasks.append(task)
-                        
-                        # Update waiting time in monitor
-                        if self.monitor:
-                            wait_time = time.time() - enqueue_time
-                            self.monitor.update_task(
-                                task_id, 
-                                wait_time=wait_time,
-                                status=CrawlStatus.IN_PROGRESS
-                            )
+                # If memory pressure is low, greedily fill all available slots
+                if not self.memory_pressure_mode:
+                    slots = self.max_session_permit - len(active_tasks)
+                    while slots > 0:
+                        try:
+                            # Use get_nowait() to immediately get tasks without blocking
+                            priority, (url, task_id, retry_count, enqueue_time) = self.task_queue.get_nowait()
                             
-                    except asyncio.TimeoutError:
-                        # No tasks in queue, that's fine
-                        pass
+                            # Create and start the task
+                            task = asyncio.create_task(
+                                self.crawl_url(url, config, task_id, retry_count)
+                            )
+                            active_tasks.append(task)
+                            
+                            # Update waiting time in monitor
+                            if self.monitor:
+                                wait_time = time.time() - enqueue_time
+                                self.monitor.update_task(
+                                    task_id, 
+                                    wait_time=wait_time,
+                                    status=CrawlStatus.IN_PROGRESS
+                                )
+                            
+                            slots -= 1
+                            
+                        except asyncio.QueueEmpty:
+                            # No more tasks in queue, exit the loop
+                            break
                         
                 # Wait for completion even if queue is starved
                 if active_tasks:
@@ -470,7 +531,7 @@ class MemoryAdaptiveDispatcher(BaseDispatcher):
         self,
         urls: List[str],
         crawler: AsyncWebCrawler,
-        config: CrawlerRunConfig,
+        config: Union[CrawlerRunConfig, List[CrawlerRunConfig]],
     ) -> AsyncGenerator[CrawlerTaskResult, None]:
         self.crawler = crawler
         
@@ -500,32 +561,34 @@ class MemoryAdaptiveDispatcher(BaseDispatcher):
                         for t in active_tasks:
                             t.cancel()
                         raise exc
-                # If memory pressure is low, start new tasks
-                if not self.memory_pressure_mode and len(active_tasks) < self.max_session_permit:
-                    try:
-                        # Try to get a task with timeout
-                        priority, (url, task_id, retry_count, enqueue_time) = await asyncio.wait_for(
-                            self.task_queue.get(), timeout=0.1
-                        )
-                        
-                        # Create and start the task
-                        task = asyncio.create_task(
-                            self.crawl_url(url, config, task_id, retry_count)
-                        )
-                        active_tasks.append(task)
-                        
-                        # Update waiting time in monitor
-                        if self.monitor:
-                            wait_time = time.time() - enqueue_time
-                            self.monitor.update_task(
-                                task_id, 
-                                wait_time=wait_time,
-                                status=CrawlStatus.IN_PROGRESS
-                            )
+                # If memory pressure is low, greedily fill all available slots
+                if not self.memory_pressure_mode:
+                    slots = self.max_session_permit - len(active_tasks)
+                    while slots > 0:
+                        try:
+                            # Use get_nowait() to immediately get tasks without blocking
+                            priority, (url, task_id, retry_count, enqueue_time) = self.task_queue.get_nowait()
                             
-                    except asyncio.TimeoutError:
-                        # No tasks in queue, that's fine
-                        pass
+                            # Create and start the task
+                            task = asyncio.create_task(
+                                self.crawl_url(url, config, task_id, retry_count)
+                            )
+                            active_tasks.append(task)
+                            
+                            # Update waiting time in monitor
+                            if self.monitor:
+                                wait_time = time.time() - enqueue_time
+                                self.monitor.update_task(
+                                    task_id, 
+                                    wait_time=wait_time,
+                                    status=CrawlStatus.IN_PROGRESS
+                                )
+                            
+                            slots -= 1
+                            
+                        except asyncio.QueueEmpty:
+                            # No more tasks in queue, exit the loop
+                            break
                         
                 # Process completed tasks and yield results
                 if active_tasks:
@@ -572,7 +635,7 @@ class SemaphoreDispatcher(BaseDispatcher):
     async def crawl_url(
         self,
         url: str,
-        config: CrawlerRunConfig,
+        config: Union[CrawlerRunConfig, List[CrawlerRunConfig]],
         task_id: str,
         semaphore: asyncio.Semaphore = None,
     ) -> CrawlerTaskResult:
@@ -580,6 +643,36 @@ class SemaphoreDispatcher(BaseDispatcher):
         error_message = ""
         memory_usage = peak_memory = 0.0
 
+        # Select appropriate config for this URL
+        selected_config = self.select_config(url, config)
+        
+        # If no config matches, return failed result
+        if selected_config is None:
+            error_message = f"No matching configuration found for URL: {url}"
+            if self.monitor:
+                self.monitor.update_task(
+                    task_id, 
+                    status=CrawlStatus.FAILED,
+                    error_message=error_message
+                )
+            
+            return CrawlerTaskResult(
+                task_id=task_id,
+                url=url,
+                result=CrawlResult(
+                    url=url, 
+                    html="", 
+                    metadata={"status": "no_config_match"}, 
+                    success=False, 
+                    error_message=error_message
+                ),
+                memory_usage=0,
+                peak_memory=0,
+                start_time=start_time,
+                end_time=time.time(),
+                error_message=error_message
+            )
+
         try:
             if self.monitor:
                 self.monitor.update_task(
@@ -592,7 +685,7 @@ class SemaphoreDispatcher(BaseDispatcher):
             async with semaphore:
                 process = psutil.Process()
                 start_memory = process.memory_info().rss / (1024 * 1024)
-                result = await self.crawler.arun(url, config=config, session_id=task_id)
+                result = await self.crawler.arun(url, config=selected_config, session_id=task_id)
                 end_memory = process.memory_info().rss / (1024 * 1024)
 
                 memory_usage = peak_memory = end_memory - start_memory
@@ -654,7 +747,7 @@ class SemaphoreDispatcher(BaseDispatcher):
         self,
         crawler: AsyncWebCrawler,  # noqa: F821
         urls: List[str],
-        config: CrawlerRunConfig,
+        config: Union[CrawlerRunConfig, List[CrawlerRunConfig]],
     ) -> List[CrawlerTaskResult]:
         self.crawler = crawler
         if self.monitor:

crawl4ai/async_url_seeder.py

@@ -829,7 +829,7 @@ class AsyncUrlSeeder:
 
     async def _iter_sitemap(self, url: str):
         try:
-            r = await self.client.get(url, timeout=15)
+            r = await self.client.get(url, timeout=15, follow_redirects=True)
             r.raise_for_status()
         except httpx.HTTPStatusError as e:
             self._log("warning", "Failed to fetch sitemap {url}: HTTP {status_code}",

crawl4ai/async_webcrawler.py

@@ -502,9 +502,12 @@ class AsyncWebCrawler:
             metadata = result.get("metadata", {})
         else:
             cleaned_html = sanitize_input_encode(result.cleaned_html)
-            media = result.media.model_dump()
-            tables = media.pop("tables", [])
-            links = result.links.model_dump()
+            # media = result.media.model_dump()
+            # tables = media.pop("tables", [])
+            # links = result.links.model_dump()
+            media = result.media.model_dump() if hasattr(result.media, 'model_dump') else result.media
+            tables = media.pop("tables", []) if isinstance(media, dict) else []
+            links = result.links.model_dump() if hasattr(result.links, 'model_dump') else result.links
             metadata = result.metadata
 
         fit_html = preprocess_html_for_schema(html_content=html, text_threshold= 500, max_size= 300_000)
@@ -650,7 +653,7 @@ class AsyncWebCrawler:
     async def arun_many(
         self,
         urls: List[str],
-        config: Optional[CrawlerRunConfig] = None,
+        config: Optional[Union[CrawlerRunConfig, List[CrawlerRunConfig]]] = None,
         dispatcher: Optional[BaseDispatcher] = None,
         # Legacy parameters maintained for backwards compatibility
         # word_count_threshold=MIN_WORD_THRESHOLD,
@@ -671,7 +674,9 @@ class AsyncWebCrawler:
 
         Args:
         urls: List of URLs to crawl
-        config: Configuration object controlling crawl behavior for all URLs
+        config: Configuration object(s) controlling crawl behavior. Can be:
+            - Single CrawlerRunConfig: Used for all URLs
+            - List[CrawlerRunConfig]: Configs with url_matcher for URL-specific settings
         dispatcher: The dispatcher strategy instance to use. Defaults to MemoryAdaptiveDispatcher
         [other parameters maintained for backwards compatibility]
 
@@ -736,7 +741,11 @@ class AsyncWebCrawler:
                 or task_result.result
             )
 
-        stream = config.stream
+        # Handle stream setting - use first config's stream setting if config is a list
+        if isinstance(config, list):
+            stream = config[0].stream if config else False
+        else:
+            stream = config.stream
 
         if stream:
 

crawl4ai/browser_adapter.py

@@ -0,0 +1,293 @@
+# browser_adapter.py
+"""
+Browser adapter for Crawl4AI to support both Playwright and undetected browsers
+with minimal changes to existing codebase.
+"""
+
+from abc import ABC, abstractmethod
+from typing import List, Dict, Any, Optional, Callable
+import time
+import json
+
+# Import both, but use conditionally
+try:
+    from playwright.async_api import Page
+except ImportError:
+    Page = Any
+
+try:
+    from patchright.async_api import Page as UndetectedPage
+except ImportError:
+    UndetectedPage = Any
+
+
+class BrowserAdapter(ABC):
+    """Abstract adapter for browser-specific operations"""
+    
+    @abstractmethod
+    async def evaluate(self, page: Page, expression: str, arg: Any = None) -> Any:
+        """Execute JavaScript in the page"""
+        pass
+    
+    @abstractmethod
+    async def setup_console_capture(self, page: Page, captured_console: List[Dict]) -> Optional[Callable]:
+        """Setup console message capturing, returns handler function if needed"""
+        pass
+    
+    @abstractmethod
+    async def setup_error_capture(self, page: Page, captured_console: List[Dict]) -> Optional[Callable]:
+        """Setup error capturing, returns handler function if needed"""
+        pass
+    
+    @abstractmethod
+    async def retrieve_console_messages(self, page: Page) -> List[Dict]:
+        """Retrieve captured console messages (for undetected browsers)"""
+        pass
+    
+    @abstractmethod
+    async def cleanup_console_capture(self, page: Page, handle_console: Optional[Callable], handle_error: Optional[Callable]):
+        """Clean up console event listeners"""
+        pass
+    
+    @abstractmethod
+    def get_imports(self) -> tuple:
+        """Get the appropriate imports for this adapter"""
+        pass
+
+
+class PlaywrightAdapter(BrowserAdapter):
+    """Adapter for standard Playwright"""
+    
+    async def evaluate(self, page: Page, expression: str, arg: Any = None) -> Any:
+        """Standard Playwright evaluate"""
+        if arg is not None:
+            return await page.evaluate(expression, arg)
+        return await page.evaluate(expression)
+    
+    async def setup_console_capture(self, page: Page, captured_console: List[Dict]) -> Optional[Callable]:
+        """Setup console capture using Playwright's event system"""
+        def handle_console_capture(msg):
+            try:
+                message_type = "unknown"
+                try:
+                    message_type = msg.type
+                except:
+                    pass
+                    
+                message_text = "unknown"
+                try:
+                    message_text = msg.text
+                except:
+                    pass
+                    
+                entry = {
+                    "type": message_type,
+                    "text": message_text,
+                    "timestamp": time.time()
+                }
+                
+                captured_console.append(entry)
+                
+            except Exception as e:
+                captured_console.append({
+                    "type": "console_capture_error", 
+                    "error": str(e), 
+                    "timestamp": time.time()
+                })
+        
+        page.on("console", handle_console_capture)
+        return handle_console_capture
+    
+    async def setup_error_capture(self, page: Page, captured_console: List[Dict]) -> Optional[Callable]:
+        """Setup error capture using Playwright's event system"""
+        def handle_pageerror_capture(err):
+            try:
+                error_message = "Unknown error"
+                try:
+                    error_message = err.message
+                except:
+                    pass
+                    
+                error_stack = ""
+                try:
+                    error_stack = err.stack
+                except:
+                    pass
+                    
+                captured_console.append({
+                    "type": "error",
+                    "text": error_message,
+                    "stack": error_stack,
+                    "timestamp": time.time()
+                })
+            except Exception as e:
+                captured_console.append({
+                    "type": "pageerror_capture_error", 
+                    "error": str(e), 
+                    "timestamp": time.time()
+                })
+        
+        page.on("pageerror", handle_pageerror_capture)
+        return handle_pageerror_capture
+    
+    async def retrieve_console_messages(self, page: Page) -> List[Dict]:
+        """Not needed for Playwright - messages are captured via events"""
+        return []
+    
+    async def cleanup_console_capture(self, page: Page, handle_console: Optional[Callable], handle_error: Optional[Callable]):
+        """Remove event listeners"""
+        if handle_console:
+            page.remove_listener("console", handle_console)
+        if handle_error:
+            page.remove_listener("pageerror", handle_error)
+    
+    def get_imports(self) -> tuple:
+        """Return Playwright imports"""
+        from playwright.async_api import Page, Error
+        from playwright.async_api import TimeoutError as PlaywrightTimeoutError
+        return Page, Error, PlaywrightTimeoutError
+
+
+class UndetectedAdapter(BrowserAdapter):
+    """Adapter for undetected browser automation with stealth features"""
+    
+    def __init__(self):
+        self._console_script_injected = {}
+    
+    async def evaluate(self, page: UndetectedPage, expression: str, arg: Any = None) -> Any:
+        """Undetected browser evaluate with isolated context"""
+        # For most evaluations, use isolated context for stealth
+        # Only use non-isolated when we need to access our injected console capture
+        isolated = not (
+            "__console" in expression or 
+            "__captured" in expression or
+            "__error" in expression or
+            "window.__" in expression
+        )
+        
+        if arg is not None:
+            return await page.evaluate(expression, arg, isolated_context=isolated)
+        return await page.evaluate(expression, isolated_context=isolated)
+    
+    async def setup_console_capture(self, page: UndetectedPage, captured_console: List[Dict]) -> Optional[Callable]:
+        """Setup console capture using JavaScript injection for undetected browsers"""
+        if not self._console_script_injected.get(page, False):
+            await page.add_init_script("""
+                // Initialize console capture
+                window.__capturedConsole = [];
+                window.__capturedErrors = [];
+                
+                // Store original console methods
+                const originalConsole = {};
+                ['log', 'info', 'warn', 'error', 'debug'].forEach(method => {
+                    originalConsole[method] = console[method];
+                    console[method] = function(...args) {
+                        try {
+                            window.__capturedConsole.push({
+                                type: method,
+                                text: args.map(arg => {
+                                    try {
+                                        if (typeof arg === 'object') {
+                                            return JSON.stringify(arg);
+                                        }
+                                        return String(arg);
+                                    } catch (e) {
+                                        return '[Object]';
+                                    }
+                                }).join(' '),
+                                timestamp: Date.now()
+                            });
+                        } catch (e) {
+                            // Fail silently to avoid detection
+                        }
+                        
+                        // Call original method
+                        originalConsole[method].apply(console, args);
+                    };
+                });
+            """)
+            self._console_script_injected[page] = True
+        
+        return None  # No handler function needed for undetected browser
+    
+    async def setup_error_capture(self, page: UndetectedPage, captured_console: List[Dict]) -> Optional[Callable]:
+        """Setup error capture using JavaScript injection for undetected browsers"""
+        if not self._console_script_injected.get(page, False):
+            await page.add_init_script("""
+                // Capture errors
+                window.addEventListener('error', (event) => {
+                    try {
+                        window.__capturedErrors.push({
+                            type: 'error',
+                            text: event.message,
+                            stack: event.error ? event.error.stack : '',
+                            filename: event.filename,
+                            lineno: event.lineno,
+                            colno: event.colno,
+                            timestamp: Date.now()
+                        });
+                    } catch (e) {
+                        // Fail silently
+                    }
+                });
+                
+                // Capture unhandled promise rejections
+                window.addEventListener('unhandledrejection', (event) => {
+                    try {
+                        window.__capturedErrors.push({
+                            type: 'unhandledrejection',
+                            text: event.reason ? String(event.reason) : 'Unhandled Promise Rejection',
+                            stack: event.reason && event.reason.stack ? event.reason.stack : '',
+                            timestamp: Date.now()
+                        });
+                    } catch (e) {
+                        // Fail silently
+                    }
+                });
+            """)
+            self._console_script_injected[page] = True
+        
+        return None  # No handler function needed for undetected browser
+    
+    async def retrieve_console_messages(self, page: UndetectedPage) -> List[Dict]:
+        """Retrieve captured console messages and errors from the page"""
+        messages = []
+        
+        try:
+            # Get console messages
+            console_messages = await page.evaluate(
+                "() => { const msgs = window.__capturedConsole || []; window.__capturedConsole = []; return msgs; }",
+                isolated_context=False
+            )
+            messages.extend(console_messages)
+            
+            # Get errors
+            errors = await page.evaluate(
+                "() => { const errs = window.__capturedErrors || []; window.__capturedErrors = []; return errs; }",
+                isolated_context=False
+            )
+            messages.extend(errors)
+            
+            # Convert timestamps from JS to Python format
+            for msg in messages:
+                if 'timestamp' in msg and isinstance(msg['timestamp'], (int, float)):
+                    msg['timestamp'] = msg['timestamp'] / 1000.0  # Convert from ms to seconds
+                    
+        except Exception:
+            # If retrieval fails, return empty list
+            pass
+        
+        return messages
+    
+    async def cleanup_console_capture(self, page: UndetectedPage, handle_console: Optional[Callable], handle_error: Optional[Callable]):
+        """Clean up for undetected browser - retrieve final messages"""
+        # For undetected browser, we don't have event listeners to remove
+        # but we should retrieve any final messages
+        final_messages = await self.retrieve_console_messages(page)
+        return final_messages
+    
+    def get_imports(self) -> tuple:
+        """Return undetected browser imports"""
+        from patchright.async_api import Page, Error
+        from patchright.async_api import TimeoutError as PlaywrightTimeoutError
+        return Page, Error, PlaywrightTimeoutError

crawl4ai/browser_manager.py

@@ -573,21 +573,26 @@ class BrowserManager:
     _playwright_instance = None
     
     @classmethod
-    async def get_playwright(cls):
-        from playwright.async_api import async_playwright
+    async def get_playwright(cls, use_undetected: bool = False):
+        if use_undetected:
+            from patchright.async_api import async_playwright
+        else:
+            from playwright.async_api import async_playwright
         cls._playwright_instance = await async_playwright().start()
         return cls._playwright_instance    
 
-    def __init__(self, browser_config: BrowserConfig, logger=None):
+    def __init__(self, browser_config: BrowserConfig, logger=None, use_undetected: bool = False):
         """
         Initialize the BrowserManager with a browser configuration.
 
         Args:
             browser_config (BrowserConfig): Configuration object containing all browser settings
             logger: Logger instance for recording events and errors
+            use_undetected (bool): Whether to use undetected browser (Patchright)
         """
         self.config: BrowserConfig = browser_config
         self.logger = logger
+        self.use_undetected = use_undetected
 
         # Browser state
         self.browser = None
@@ -601,7 +606,16 @@ class BrowserManager:
 
         # Keep track of contexts by a "config signature," so each unique config reuses a single context
         self.contexts_by_config = {}
-        self._contexts_lock = asyncio.Lock() 
+        self._contexts_lock = asyncio.Lock()
+        
+        # Serialize context.new_page() across concurrent tasks to avoid races
+        # when using a shared persistent context (context.pages may be empty
+        # for all racers). Prevents 'Target page/context closed' errors.
+        self._page_lock = asyncio.Lock()
+        
+        # Stealth-related attributes
+        self._stealth_instance = None
+        self._stealth_cm = None 
 
         # Initialize ManagedBrowser if needed
         if self.config.use_managed_browser:
@@ -630,9 +644,21 @@ class BrowserManager:
         if self.playwright is not None:
             await self.close()
             
-        from playwright.async_api import async_playwright
+        if self.use_undetected:
+            from patchright.async_api import async_playwright
+        else:
+            from playwright.async_api import async_playwright
 
-        self.playwright = await async_playwright().start()
+        # Initialize playwright with or without stealth
+        if self.config.enable_stealth and not self.use_undetected:
+            # Import stealth only when needed
+            from playwright_stealth import Stealth
+            # Use the recommended stealth wrapper approach
+            self._stealth_instance = Stealth()
+            self._stealth_cm = self._stealth_instance.use_async(async_playwright())
+            self.playwright = await self._stealth_cm.__aenter__()
+        else:
+            self.playwright = await async_playwright().start()
 
         if self.config.cdp_url or self.config.use_managed_browser:
             self.config.use_managed_browser = True
@@ -1006,13 +1032,26 @@ class BrowserManager:
                 context = await self.create_browser_context(crawlerRunConfig)
                 ctx = self.default_context        # default context, one window only
                 ctx = await clone_runtime_state(context, ctx, crawlerRunConfig, self.config)
-                page = await ctx.new_page()
+                # Avoid concurrent new_page on shared persistent context
+                # See GH-1198: context.pages can be empty under races
+                async with self._page_lock:
+                    page = await ctx.new_page()
             else:
                 context = self.default_context
                 pages = context.pages
                 page = next((p for p in pages if p.url == crawlerRunConfig.url), None)
                 if not page:
-                    page = context.pages[0] # await context.new_page()
+                    if pages:
+                        page = pages[0]
+                    else:
+                        # Double-check under lock to avoid TOCTOU and ensure only
+                        # one task calls new_page when pages=[] concurrently
+                        async with self._page_lock:
+                            pages = context.pages
+                            if pages:
+                                page = pages[0]
+                            else:
+                                page = await context.new_page()
         else:
             # Otherwise, check if we have an existing context for this config
             config_signature = self._make_config_signature(crawlerRunConfig)
@@ -1094,5 +1133,19 @@ class BrowserManager:
             self.managed_browser = None
 
         if self.playwright:
-            await self.playwright.stop()
+            # Handle stealth context manager cleanup if it exists
+            if hasattr(self, '_stealth_cm') and self._stealth_cm is not None:
+                try:
+                    await self._stealth_cm.__aexit__(None, None, None)
+                except Exception as e:
+                    if self.logger:
+                        self.logger.error(
+                            message="Error closing stealth context: {error}",
+                            tag="ERROR", 
+                            params={"error": str(e)}
+                        )
+                self._stealth_cm = None
+                self._stealth_instance = None
+            else:
+                await self.playwright.stop()
             self.playwright = None

crawl4ai/browser_profiler.py

@@ -65,6 +65,213 @@ class BrowserProfiler:
         self.builtin_config_file = os.path.join(self.builtin_browser_dir, "browser_config.json")
         os.makedirs(self.builtin_browser_dir, exist_ok=True)
     
+    def _is_windows(self) -> bool:
+        """Check if running on Windows platform."""
+        return sys.platform.startswith('win') or sys.platform == 'cygwin'
+    
+    def _is_macos(self) -> bool:
+        """Check if running on macOS platform."""
+        return sys.platform == 'darwin'
+    
+    def _is_linux(self) -> bool:
+        """Check if running on Linux platform."""
+        return sys.platform.startswith('linux')
+    
+    def _get_quit_message(self, tag: str) -> str:
+        """Get appropriate quit message based on context."""
+        if tag == "PROFILE":
+            return "Closing browser and saving profile..."
+        elif tag == "CDP":
+            return "Closing browser..."
+        else:
+            return "Closing browser..."
+    
+    async def _listen_windows(self, user_done_event, check_browser_process, tag: str):
+        """Windows-specific keyboard listener using msvcrt."""
+        try:
+            import msvcrt
+        except ImportError:
+            raise ImportError("msvcrt module not available on this platform")
+        
+        while True:
+            try:
+                # Check for keyboard input
+                if msvcrt.kbhit():
+                    raw = msvcrt.getch()
+                    
+                    # Handle Unicode decoding more robustly
+                    key = None
+                    try:
+                        key = raw.decode("utf-8")
+                    except UnicodeDecodeError:
+                        try:
+                            # Try different encodings
+                            key = raw.decode("latin1")
+                        except UnicodeDecodeError:
+                            # Skip if we can't decode
+                            continue
+                    
+                    # Validate key
+                    if not key or len(key) != 1:
+                        continue
+                    
+                    # Check for printable characters only
+                    if not key.isprintable():
+                        continue
+                    
+                    # Check for quit command
+                    if key.lower() == "q":
+                        self.logger.info(
+                            self._get_quit_message(tag),
+                            tag=tag,
+                            base_color=LogColor.GREEN
+                        )
+                        user_done_event.set()
+                        return
+                
+                # Check if browser process ended
+                if await check_browser_process():
+                    return
+                
+                # Small delay to prevent busy waiting
+                await asyncio.sleep(0.1)
+                
+            except Exception as e:
+                self.logger.warning(f"Error in Windows keyboard listener: {e}", tag=tag)
+                # Continue trying instead of failing completely
+                await asyncio.sleep(0.1)
+                continue
+    
+    async def _listen_unix(self, user_done_event: asyncio.Event, check_browser_process, tag: str):
+        """Unix/Linux/macOS keyboard listener using termios and select."""
+        try:
+            import termios
+            import tty
+            import select
+        except ImportError:
+            raise ImportError("termios/tty/select modules not available on this platform")
+        
+        # Get stdin file descriptor
+        try:
+            fd = sys.stdin.fileno()
+        except (AttributeError, OSError):
+            raise ImportError("stdin is not a terminal")
+        
+        # Save original terminal settings
+        old_settings = None
+        try:
+            old_settings = termios.tcgetattr(fd)
+        except termios.error as e:
+            raise ImportError(f"Cannot get terminal attributes: {e}")
+        
+        try:
+            # Switch to non-canonical mode (cbreak mode)
+            tty.setcbreak(fd)
+            
+            while True:
+                try:
+                    # Use select to check if input is available (non-blocking)
+                    # Timeout of 0.5 seconds to periodically check browser process
+                    readable, _, _ = select.select([sys.stdin], [], [], 0.5)
+                    
+                    if readable:
+                        # Read one character
+                        key = sys.stdin.read(1)
+                        
+                        if key and key.lower() == "q":
+                            self.logger.info(
+                                self._get_quit_message(tag),
+                                tag=tag,
+                                base_color=LogColor.GREEN
+                            )
+                            user_done_event.set()
+                            return
+                    
+                    # Check if browser process ended
+                    if await check_browser_process():
+                        return
+                    
+                    # Small delay to prevent busy waiting
+                    await asyncio.sleep(0.1)
+                    
+                except (KeyboardInterrupt, EOFError):
+                    # Handle Ctrl+C or EOF gracefully
+                    self.logger.info("Keyboard interrupt received", tag=tag)
+                    user_done_event.set()
+                    return
+                except Exception as e:
+                    self.logger.warning(f"Error in Unix keyboard listener: {e}", tag=tag)
+                    await asyncio.sleep(0.1)
+                    continue
+                    
+        finally:
+            # Always restore terminal settings
+            if old_settings is not None:
+                try:
+                    termios.tcsetattr(fd, termios.TCSADRAIN, old_settings)
+                except Exception as e:
+                    self.logger.error(f"Failed to restore terminal settings: {e}", tag=tag)
+    
+    async def _listen_fallback(self, user_done_event: asyncio.Event, check_browser_process, tag: str):
+        """Fallback keyboard listener using simple input() method."""
+        self.logger.info("Using fallback input mode. Type 'q' and press Enter to quit.", tag=tag)
+        
+        # Run input in a separate thread to avoid blocking
+        import threading
+        import queue
+        
+        input_queue = queue.Queue()
+        
+        def input_thread():
+            """Thread function to handle input."""
+            try:
+                while not user_done_event.is_set():
+                    try:
+                        # Use input() with a prompt
+                        user_input = input("Press 'q' + Enter to quit: ").strip().lower()
+                        input_queue.put(user_input)
+                        if user_input == 'q':
+                            break
+                    except (EOFError, KeyboardInterrupt):
+                        input_queue.put('q')
+                        break
+                    except Exception as e:
+                        self.logger.warning(f"Error in input thread: {e}", tag=tag)
+                        break
+            except Exception as e:
+                self.logger.error(f"Input thread failed: {e}", tag=tag)
+        
+        # Start input thread
+        thread = threading.Thread(target=input_thread, daemon=True)
+        thread.start()
+        
+        try:
+            while not user_done_event.is_set():
+                # Check for user input
+                try:
+                    user_input = input_queue.get_nowait()
+                    if user_input == 'q':
+                        self.logger.info(
+                            self._get_quit_message(tag),
+                            tag=tag,
+                            base_color=LogColor.GREEN
+                        )
+                        user_done_event.set()
+                        return
+                except queue.Empty:
+                    pass
+                
+                # Check if browser process ended
+                if await check_browser_process():
+                    return
+                
+                # Small delay
+                await asyncio.sleep(0.5)
+                
+        except Exception as e:
+            self.logger.error(f"Fallback listener failed: {e}", tag=tag)
+            user_done_event.set()
+    
     async def create_profile(self, 
                             profile_name: Optional[str] = None, 
                             browser_config: Optional[BrowserConfig] = None) -> Optional[str]:
@@ -180,42 +387,38 @@ class BrowserProfiler:
         
         # Run keyboard input loop in a separate task
         async def listen_for_quit_command():
-            import termios
-            import tty
-            import select
-            
+            """Cross-platform keyboard listener that waits for 'q' key press."""
             # First output the prompt
-            self.logger.info("Press 'q' when you've finished using the browser...", tag="PROFILE")
-            
-            # Save original terminal settings
-            fd = sys.stdin.fileno()
-            old_settings = termios.tcgetattr(fd)
-            
+            self.logger.info(
+                "Press {segment} when you've finished using the browser...",
+                tag="PROFILE",
+                params={"segment": "'q'"}, colors={"segment": LogColor.YELLOW},
+                base_color=LogColor.CYAN
+            )
+
+            async def check_browser_process():
+                """Check if browser process is still running."""
+                if (
+                    managed_browser.browser_process
+                    and managed_browser.browser_process.poll() is not None
+                ):
+                    self.logger.info(
+                        "Browser already closed. Ending input listener.", tag="PROFILE"
+                    )
+                    user_done_event.set()
+                    return True
+                return False
+
+            # Try platform-specific implementations with fallback
             try:
-                # Switch to non-canonical mode (no line buffering)
-                tty.setcbreak(fd)
-                
-                while True:
-                    # Check if input is available (non-blocking)
-                    readable, _, _ = select.select([sys.stdin], [], [], 0.5)
-                    if readable:
-                        key = sys.stdin.read(1)
-                        if key.lower() == 'q':
-                            self.logger.info("Closing browser and saving profile...", tag="PROFILE", base_color=LogColor.GREEN)
-                            user_done_event.set()
-                            return
-                    
-                    # Check if the browser process has already exited
-                    if managed_browser.browser_process and managed_browser.browser_process.poll() is not None:
-                        self.logger.info("Browser already closed. Ending input listener.", tag="PROFILE")
-                        user_done_event.set()
-                        return
-                        
-                    await asyncio.sleep(0.1)
-            
-            finally:
-                # Restore terminal settings 
-                termios.tcsetattr(fd, termios.TCSADRAIN, old_settings)
+                if self._is_windows():
+                    await self._listen_windows(user_done_event, check_browser_process, "PROFILE")
+                else:
+                    await self._listen_unix(user_done_event, check_browser_process, "PROFILE")
+            except Exception as e:
+                self.logger.warning(f"Platform-specific keyboard listener failed: {e}", tag="PROFILE")
+                self.logger.info("Falling back to simple input mode...", tag="PROFILE")
+                await self._listen_fallback(user_done_event, check_browser_process, "PROFILE")
         
         try:
             from playwright.async_api import async_playwright
@@ -682,42 +885,33 @@ class BrowserProfiler:
         
         # Run keyboard input loop in a separate task
         async def listen_for_quit_command():
-            import termios
-            import tty
-            import select
-            
+            """Cross-platform keyboard listener that waits for 'q' key press."""
             # First output the prompt
-            self.logger.info("Press 'q' to stop the browser and exit...", tag="CDP")
-            
-            # Save original terminal settings
-            fd = sys.stdin.fileno()
-            old_settings = termios.tcgetattr(fd)
-            
+            self.logger.info(
+                "Press {segment} to stop the browser and exit...",
+                tag="CDP",
+                params={"segment": "'q'"}, colors={"segment": LogColor.YELLOW},
+                base_color=LogColor.CYAN
+            )
+
+            async def check_browser_process():
+                """Check if browser process is still running."""
+                if managed_browser.browser_process and managed_browser.browser_process.poll() is not None:
+                    self.logger.info("Browser already closed. Ending input listener.", tag="CDP")
+                    user_done_event.set()
+                    return True
+                return False
+
+            # Try platform-specific implementations with fallback
             try:
-                # Switch to non-canonical mode (no line buffering)
-                tty.setcbreak(fd)
-                
-                while True:
-                    # Check if input is available (non-blocking)
-                    readable, _, _ = select.select([sys.stdin], [], [], 0.5)
-                    if readable:
-                        key = sys.stdin.read(1)
-                        if key.lower() == 'q':
-                            self.logger.info("Closing browser...", tag="CDP")
-                            user_done_event.set()
-                            return
-                    
-                    # Check if the browser process has already exited
-                    if managed_browser.browser_process and managed_browser.browser_process.poll() is not None:
-                        self.logger.info("Browser already closed. Ending input listener.", tag="CDP")
-                        user_done_event.set()
-                        return
-                        
-                    await asyncio.sleep(0.1)
-            
-            finally:
-                # Restore terminal settings 
-                termios.tcsetattr(fd, termios.TCSADRAIN, old_settings)
+                if self._is_windows():
+                    await self._listen_windows(user_done_event, check_browser_process, "CDP")
+                else:
+                    await self._listen_unix(user_done_event, check_browser_process, "CDP")
+            except Exception as e:
+                self.logger.warning(f"Platform-specific keyboard listener failed: {e}", tag="CDP")
+                self.logger.info("Falling back to simple input mode...", tag="CDP")
+                await self._listen_fallback(user_done_event, check_browser_process, "CDP")
                 
         # Function to retrieve and display CDP JSON config
         async def get_cdp_json(port):

crawl4ai/cli.py

@@ -27,7 +27,10 @@ from crawl4ai import (
     PruningContentFilter,
     BrowserProfiler,
     DefaultMarkdownGenerator,
-    LLMConfig
+    LLMConfig,
+    BFSDeepCrawlStrategy,
+    DFSDeepCrawlStrategy,
+    BestFirstCrawlingStrategy,
 )
 from crawl4ai.config import USER_SETTINGS
 from litellm import completion
@@ -1014,9 +1017,11 @@ def cdp_cmd(user_data_dir: Optional[str], port: int, browser_type: str, headless
 @click.option("--question", "-q", help="Ask a question about the crawled content")
 @click.option("--verbose", "-v", is_flag=True)
 @click.option("--profile", "-p", help="Use a specific browser profile (by name)")
+@click.option("--deep-crawl", type=click.Choice(["bfs", "dfs", "best-first"]), help="Enable deep crawling with specified strategy (bfs, dfs, or best-first)")
+@click.option("--max-pages", type=int, default=10, help="Maximum number of pages to crawl in deep crawl mode")
 def crawl_cmd(url: str, browser_config: str, crawler_config: str, filter_config: str, 
            extraction_config: str, json_extract: str, schema: str, browser: Dict, crawler: Dict,
-           output: str, output_file: str, bypass_cache: bool, question: str, verbose: bool, profile: str):
+           output: str, output_file: str, bypass_cache: bool, question: str, verbose: bool, profile: str, deep_crawl: str, max_pages: int):
     """Crawl a website and extract content
     
     Simple Usage:
@@ -1156,6 +1161,27 @@ Always return valid, properly formatted JSON."""
 
         crawler_cfg.scraping_strategy = LXMLWebScrapingStrategy()    
 
+        # Handle deep crawling configuration
+        if deep_crawl:
+            if deep_crawl == "bfs":
+                crawler_cfg.deep_crawl_strategy = BFSDeepCrawlStrategy(
+                    max_depth=3,
+                    max_pages=max_pages
+                )
+            elif deep_crawl == "dfs":
+                crawler_cfg.deep_crawl_strategy = DFSDeepCrawlStrategy(
+                    max_depth=3,
+                    max_pages=max_pages
+                )
+            elif deep_crawl == "best-first":
+                crawler_cfg.deep_crawl_strategy = BestFirstCrawlingStrategy(
+                    max_depth=3,
+                    max_pages=max_pages
+                )
+            
+            if verbose:
+                console.print(f"[green]Deep crawling enabled:[/green] {deep_crawl} strategy, max {max_pages} pages")
+
         config = get_global_config()
         
         browser_cfg.verbose = config.get("VERBOSE", False)
@@ -1170,39 +1196,60 @@ Always return valid, properly formatted JSON."""
             verbose
         )
 
+        # Handle deep crawl results (list) vs single result
+        if isinstance(result, list):
+            if len(result) == 0:
+                click.echo("No results found during deep crawling")
+                return
+            # Use the first result for question answering and output
+            main_result = result[0]
+            all_results = result
+        else:
+            # Single result from regular crawling
+            main_result = result
+            all_results = [result]
+
         # Handle question
         if question:
             provider, token = setup_llm_config()
-            markdown = result.markdown.raw_markdown
+            markdown = main_result.markdown.raw_markdown
             anyio.run(stream_llm_response, url, markdown, question, provider, token)
             return
         
         # Handle output
         if not output_file:
             if output == "all":
-                click.echo(json.dumps(result.model_dump(), indent=2))
+                if isinstance(result, list):
+                    output_data = [r.model_dump() for r in all_results]
+                    click.echo(json.dumps(output_data, indent=2))
+                else:
+                    click.echo(json.dumps(main_result.model_dump(), indent=2))
             elif output == "json":
-                print(result.extracted_content)
-                extracted_items = json.loads(result.extracted_content)
+                print(main_result.extracted_content)
+                extracted_items = json.loads(main_result.extracted_content)
                 click.echo(json.dumps(extracted_items, indent=2))
                 
             elif output in ["markdown", "md"]:
-                click.echo(result.markdown.raw_markdown)
+                click.echo(main_result.markdown.raw_markdown)
             elif output in ["markdown-fit", "md-fit"]:
-                click.echo(result.markdown.fit_markdown)
+                click.echo(main_result.markdown.fit_markdown)
         else:
             if output == "all":
                 with open(output_file, "w") as f:
-                    f.write(json.dumps(result.model_dump(), indent=2))
+                    if isinstance(result, list):
+                        output_data = [r.model_dump() for r in all_results]
+                        f.write(json.dumps(output_data, indent=2))
+                    else:
+                        f.write(json.dumps(main_result.model_dump(), indent=2))
             elif output == "json":
                 with open(output_file, "w") as f:
-                    f.write(result.extracted_content)
+                    f.write(main_result.extracted_content)
             elif output in ["markdown", "md"]:
                 with open(output_file, "w") as f:
-                    f.write(result.markdown.raw_markdown)
+                    f.write(main_result.markdown.raw_markdown)
             elif output in ["markdown-fit", "md-fit"]:
                 with open(output_file, "w") as f:
-                    f.write(result.markdown.fit_markdown)
+                    f.write(main_result.markdown.fit_markdown)
             
     except Exception as e:
         raise click.ClickException(str(e))
@@ -1354,9 +1401,11 @@ def profiles_cmd():
 @click.option("--question", "-q", help="Ask a question about the crawled content")
 @click.option("--verbose", "-v", is_flag=True)
 @click.option("--profile", "-p", help="Use a specific browser profile (by name)")
+@click.option("--deep-crawl", type=click.Choice(["bfs", "dfs", "best-first"]), help="Enable deep crawling with specified strategy")
+@click.option("--max-pages", type=int, default=10, help="Maximum number of pages to crawl in deep crawl mode")
 def default(url: str, example: bool, browser_config: str, crawler_config: str, filter_config: str, 
         extraction_config: str, json_extract: str, schema: str, browser: Dict, crawler: Dict,
-        output: str, bypass_cache: bool, question: str, verbose: bool, profile: str):
+        output: str, bypass_cache: bool, question: str, verbose: bool, profile: str, deep_crawl: str, max_pages: int):
     """Crawl4AI CLI - Web content extraction tool
 
     Simple Usage:
@@ -1406,7 +1455,9 @@ def default(url: str, example: bool, browser_config: str, crawler_config: str, f
         bypass_cache=bypass_cache,
         question=question,
         verbose=verbose,
-        profile=profile
+        profile=profile,
+        deep_crawl=deep_crawl,
+        max_pages=max_pages
     )
 
 def main():

crawl4ai/content_scraping_strategy.py

@@ -98,20 +98,20 @@ class ContentScrapingStrategy(ABC):
         pass
 
 
-class WebScrapingStrategy(ContentScrapingStrategy):
+class LXMLWebScrapingStrategy(ContentScrapingStrategy):
     """
-    Class for web content scraping. Perhaps the most important class.
-
-    How it works:
-    1. Extract content from HTML using BeautifulSoup.
-    2. Clean the extracted content using a content cleaning strategy.
-    3. Filter the cleaned content using a content filtering strategy.
-    4. Generate markdown content from the filtered content.
-    5. Return the markdown content.
+    LXML-based implementation for fast web content scraping.
+    
+    This is the primary scraping strategy in Crawl4AI, providing high-performance
+    HTML parsing and content extraction using the lxml library.
+    
+    Note: WebScrapingStrategy is now an alias for this class to maintain
+    backward compatibility.
     """
-
     def __init__(self, logger=None):
         self.logger = logger
+        self.DIMENSION_REGEX = re.compile(r"(\d+)(\D*)")
+        self.BASE64_PATTERN = re.compile(r'data:image/[^;]+;base64,([^"]+)')
 
     def _log(self, level, message, tag="SCRAPE", **kwargs):
         """Helper method to safely use logger."""
@@ -132,7 +132,7 @@ class WebScrapingStrategy(ContentScrapingStrategy):
             ScrapingResult: A structured result containing the scraped content.
         """
         actual_url = kwargs.get("redirected_url", url)
-        raw_result = self._scrap(actual_url, html, is_async=False, **kwargs)
+        raw_result = self._scrap(actual_url, html, **kwargs)
         if raw_result is None:
             return ScrapingResult(
                 cleaned_html="",
@@ -196,376 +196,9 @@ class WebScrapingStrategy(ContentScrapingStrategy):
         Returns:
             ScrapingResult: A structured result containing the scraped content.
         """
-        return await asyncio.to_thread(self._scrap, url, html, **kwargs)
+        return await asyncio.to_thread(self.scrap, url, html, **kwargs)
 
-    def is_data_table(self, table: Tag, **kwargs) -> bool:
-        """
-        Determine if a table element is a data table (not a layout table).
-
-        Args:
-            table (Tag): BeautifulSoup Tag representing a table element
-            **kwargs: Additional keyword arguments including table_score_threshold
-
-        Returns:
-            bool: True if the table is a data table, False otherwise
-        """
-        score = 0
-        
-        # Check for thead and tbody
-        has_thead = len(table.select('thead')) > 0
-        has_tbody = len(table.select('tbody')) > 0
-        if has_thead:
-            score += 2
-        if has_tbody:
-            score += 1
-            
-        # Check for th elements
-        th_count = len(table.select('th'))
-        if th_count > 0:
-            score += 2
-            if has_thead or len(table.select('tr:first-child th')) > 0:
-                score += 1
-                
-        # Check for nested tables
-        if len(table.select('table')) > 0:
-            score -= 3
-            
-        # Role attribute check
-        role = table.get('role', '').lower()
-        if role in {'presentation', 'none'}:
-            score -= 3
-            
-        # Column consistency
-        rows = table.select('tr')
-        if not rows:
-            return False
-            
-        col_counts = [len(row.select('td, th')) for row in rows]
-        avg_cols = sum(col_counts) / len(col_counts)
-        variance = sum((c - avg_cols)**2 for c in col_counts) / len(col_counts)
-        if variance < 1:
-            score += 2
-            
-        # Caption and summary
-        if table.select('caption'):
-            score += 2
-        if table.has_attr('summary') and table['summary']:
-            score += 1
-            
-        # Text density
-        total_text = sum(len(cell.get_text().strip()) for row in rows for cell in row.select('td, th'))
-        total_tags = sum(1 for _ in table.descendants if isinstance(_, Tag))
-        text_ratio = total_text / (total_tags + 1e-5)
-        if text_ratio > 20:
-            score += 3
-        elif text_ratio > 10:
-            score += 2
-            
-        # Data attributes
-        data_attrs = sum(1 for attr in table.attrs if attr.startswith('data-'))
-        score += data_attrs * 0.5
-        
-        # Size check
-        if avg_cols >= 2 and len(rows) >= 2:
-            score += 2
-            
-        threshold = kwargs.get('table_score_threshold', 7)
-        return score >= threshold
-    
-    def extract_table_data(self, table: Tag) -> dict:
-        """
-        Extract structured data from a table element.
-        
-        Args:
-            table (Tag): BeautifulSoup Tag representing a table element
-            
-        Returns:
-            dict: Dictionary containing table data (headers, rows, caption, summary)
-        """
-        caption_elem = table.select_one('caption')
-        caption = caption_elem.get_text().strip() if caption_elem else ""
-        summary = table.get('summary', '').strip()
-        
-        # Extract headers with colspan handling
-        headers = []
-        thead_rows = table.select('thead tr')
-        if thead_rows:
-            header_cells = thead_rows[0].select('th')
-            for cell in header_cells:
-                text = cell.get_text().strip()
-                colspan = int(cell.get('colspan', 1))
-                headers.extend([text] * colspan)
-        else:
-            first_row = table.select('tr:first-child')
-            if first_row:
-                for cell in first_row[0].select('th, td'):
-                    text = cell.get_text().strip()
-                    colspan = int(cell.get('colspan', 1))
-                    headers.extend([text] * colspan)
-        
-        # Extract rows with colspan handling
-        rows = []
-        all_rows = table.select('tr')
-        thead = table.select_one('thead')
-        tbody_rows = []
-
-        if thead:
-            thead_rows = thead.select('tr')
-            tbody_rows = [row for row in all_rows if row not in thead_rows]
-        else:
-            if all_rows and all_rows[0].select('th'):
-                tbody_rows = all_rows[1:]
-            else:
-                tbody_rows = all_rows
-                
-        for row in tbody_rows:        
-        # for row in table.select('tr:not(:has(ancestor::thead))'):
-            row_data = []
-            for cell in row.select('td'):
-                text = cell.get_text().strip()
-                colspan = int(cell.get('colspan', 1))
-                row_data.extend([text] * colspan)
-            if row_data:
-                rows.append(row_data)
-                
-        # Align rows with headers
-        max_columns = len(headers) if headers else (max(len(row) for row in rows) if rows else 0)
-        aligned_rows = []
-        for row in rows:
-            aligned = row[:max_columns] + [''] * (max_columns - len(row))
-            aligned_rows.append(aligned)
-            
-        if not headers:
-            headers = [f"Column {i+1}" for i in range(max_columns)]
-            
-        return {
-            "headers": headers,
-            "rows": aligned_rows,
-            "caption": caption,
-            "summary": summary,
-        }
-    
-    def flatten_nested_elements(self, node):
-        """
-        Flatten nested elements in a HTML tree.
-
-        Args:
-            node (Tag): The root node of the HTML tree.
-
-        Returns:
-            Tag: The flattened HTML tree.
-        """
-        if isinstance(node, NavigableString):
-            return node
-        if (
-            len(node.contents) == 1
-            and isinstance(node.contents[0], Tag)
-            and node.contents[0].name == node.name
-        ):
-            return self.flatten_nested_elements(node.contents[0])
-        node.contents = [self.flatten_nested_elements(child) for child in node.contents]
-        return node
-
-    def find_closest_parent_with_useful_text(self, tag, **kwargs):
-        """
-        Find the closest parent with useful text.
-
-        Args:
-            tag (Tag): The starting tag to search from.
-            **kwargs: Additional keyword arguments.
-
-        Returns:
-            Tag: The closest parent with useful text, or None if not found.
-        """
-        image_description_min_word_threshold = kwargs.get(
-            "image_description_min_word_threshold", IMAGE_DESCRIPTION_MIN_WORD_THRESHOLD
-        )
-        current_tag = tag
-        while current_tag:
-            current_tag = current_tag.parent
-            # Get the text content of the parent tag
-            if current_tag:
-                text_content = current_tag.get_text(separator=" ", strip=True)
-                # Check if the text content has at least word_count_threshold
-                if len(text_content.split()) >= image_description_min_word_threshold:
-                    return text_content
-        return None
-
-    def remove_unwanted_attributes(
-        self, element, important_attrs, keep_data_attributes=False
-    ):
-        """
-        Remove unwanted attributes from an HTML element.
-
-        Args:
-            element (Tag): The HTML element to remove attributes from.
-            important_attrs (list): List of important attributes to keep.
-            keep_data_attributes (bool): Whether to keep data attributes.
-
-        Returns:
-            None
-        """
-        attrs_to_remove = []
-        for attr in element.attrs:
-            if attr not in important_attrs:
-                if keep_data_attributes:
-                    if not attr.startswith("data-"):
-                        attrs_to_remove.append(attr)
-                else:
-                    attrs_to_remove.append(attr)
-
-        for attr in attrs_to_remove:
-            del element[attr]
-
-    def process_image(self, img, url, index, total_images, **kwargs):
-        """
-        Process an image element.
-
-        How it works:
-        1. Check if the image has valid display and inside undesired html elements.
-        2. Score an image for it's usefulness.
-        3. Extract image file metadata to extract size and extension.
-        4. Generate a dictionary with the processed image information.
-        5. Return the processed image information.
-
-        Args:
-            img (Tag): The image element to process.
-            url (str): The URL of the page containing the image.
-            index (int): The index of the image in the list of images.
-            total_images (int): The total number of images in the list.
-            **kwargs: Additional keyword arguments.
-
-        Returns:
-            dict: A dictionary containing the processed image information.
-        """
-        # parse_srcset = lambda s: [{'url': u.strip().split()[0], 'width': u.strip().split()[-1].rstrip('w')
-        #                 if ' ' in u else None}
-        #                 for u in [f"http{p}" for p in s.split("http") if p]]
-
-        # Constants for checks
-        classes_to_check = frozenset(["button", "icon", "logo"])
-        tags_to_check = frozenset(["button", "input"])
-        image_formats = frozenset(["jpg", "jpeg", "png", "webp", "avif", "gif"])
-
-        # Pre-fetch commonly used attributes
-        style = img.get("style", "")
-        alt = img.get("alt", "")
-        src = img.get("src", "")
-        data_src = img.get("data-src", "")
-        srcset = img.get("srcset", "")
-        data_srcset = img.get("data-srcset", "")
-        width = img.get("width")
-        height = img.get("height")
-        parent = img.parent
-        parent_classes = parent.get("class", [])
-
-        # Quick validation checks
-        if (
-            "display:none" in style
-            or parent.name in tags_to_check
-            or any(c in cls for c in parent_classes for cls in classes_to_check)
-            or any(c in src for c in classes_to_check)
-            or any(c in alt for c in classes_to_check)
-        ):
-            return None
-
-        # Quick score calculation
-        score = 0
-        if width and width.isdigit():
-            width_val = int(width)
-            score += 1 if width_val > 150 else 0
-        if height and height.isdigit():
-            height_val = int(height)
-            score += 1 if height_val > 150 else 0
-        if alt:
-            score += 1
-        score += index / total_images < 0.5
-
-        # image_format = ''
-        # if "data:image/" in src:
-        #     image_format = src.split(',')[0].split(';')[0].split('/')[1].split(';')[0]
-        # else:
-        #     image_format = os.path.splitext(src)[1].lower().strip('.').split('?')[0]
-
-        # if image_format in ('jpg', 'png', 'webp', 'avif'):
-        #     score += 1
-
-        # Check for image format in all possible sources
-        def has_image_format(url):
-            return any(fmt in url.lower() for fmt in image_formats)
-
-        # Score for having proper image sources
-        if any(has_image_format(url) for url in [src, data_src, srcset, data_srcset]):
-            score += 1
-        if srcset or data_srcset:
-            score += 1
-        if img.find_parent("picture"):
-            score += 1
-
-        # Detect format from any available source
-        detected_format = None
-        for url in [src, data_src, srcset, data_srcset]:
-            if url:
-                format_matches = [fmt for fmt in image_formats if fmt in url.lower()]
-                if format_matches:
-                    detected_format = format_matches[0]
-                    break
-
-        if score <= kwargs.get("image_score_threshold", IMAGE_SCORE_THRESHOLD):
-            return None
-
-        # Use set for deduplication
-        unique_urls = set()
-        image_variants = []
-
-        # Generate a unique group ID for this set of variants
-        group_id = index
-
-        # Base image info template
-        base_info = {
-            "alt": alt,
-            "desc": self.find_closest_parent_with_useful_text(img, **kwargs),
-            "score": score,
-            "type": "image",
-            "group_id": group_id,  # Group ID for this set of variants
-            "format": detected_format,
-        }
-
-        # Inline function for adding variants
-        def add_variant(src, width=None):
-            if src and not src.startswith("data:") and src not in unique_urls:
-                unique_urls.add(src)
-                image_variants.append({**base_info, "src": src, "width": width})
-
-        # Process all sources
-        add_variant(src)
-        add_variant(data_src)
-
-        # Handle srcset and data-srcset in one pass
-        for attr in ("srcset", "data-srcset"):
-            if value := img.get(attr):
-                for source in parse_srcset(value):
-                    add_variant(source["url"], source["width"])
-
-        # Quick picture element check
-        if picture := img.find_parent("picture"):
-            for source in picture.find_all("source"):
-                if srcset := source.get("srcset"):
-                    for src in parse_srcset(srcset):
-                        add_variant(src["url"], src["width"])
-
-        # Framework-specific attributes in one pass
-        for attr, value in img.attrs.items():
-            if (
-                attr.startswith("data-")
-                and ("src" in attr or "srcset" in attr)
-                and "http" in value
-            ):
-                add_variant(value)
-
-        return image_variants if image_variants else None
-
-    def process_element(self, url, element: PageElement, **kwargs) -> Dict[str, Any]:
+    def process_element(self, url, element: lhtml.HtmlElement, **kwargs) -> Dict[str, Any]:
         """
         Process an HTML element.
 
@@ -577,7 +210,7 @@ class WebScrapingStrategy(ContentScrapingStrategy):
 
         Args:
             url (str): The URL of the page containing the element.
-            element (Tag): The HTML element to process.
+            element (lhtml.HtmlElement): The HTML element to process.
             **kwargs: Additional keyword arguments.
 
         Returns:
@@ -595,514 +228,6 @@ class WebScrapingStrategy(ContentScrapingStrategy):
             "external_links_dict": external_links_dict,
         }
 
-    def _process_element(
-        self,
-        url,
-        element: PageElement,
-        media: Dict[str, Any],
-        internal_links_dict: Dict[str, Any],
-        external_links_dict: Dict[str, Any],
-        **kwargs,
-    ) -> bool:
-        """
-        Process an HTML element.
-        """
-        try:
-            if isinstance(element, NavigableString):
-                if isinstance(element, Comment):
-                    element.extract()
-                return False
-
-            # if element.name == 'img':
-            #     process_image(element, url, 0, 1)
-            #     return True
-            base_domain = kwargs.get("base_domain", get_base_domain(url))
-
-            if element.name in ["script", "style", "link", "meta", "noscript"]:
-                element.decompose()
-                return False
-
-            keep_element = False
-            # Special case for table elements - always preserve structure
-            if element.name in ["tr", "td", "th"]:
-                keep_element = True
-
-            exclude_domains = kwargs.get("exclude_domains", [])
-            # exclude_social_media_domains = kwargs.get('exclude_social_media_domains', set(SOCIAL_MEDIA_DOMAINS))
-            # exclude_social_media_domains = SOCIAL_MEDIA_DOMAINS + kwargs.get('exclude_social_media_domains', [])
-            # exclude_social_media_domains = list(set(exclude_social_media_domains))
-
-            try:
-                if element.name == "a" and element.get("href"):
-                    href = element.get("href", "").strip()
-                    if not href:  # Skip empty hrefs
-                        return False
-
-                    # url_base = url.split("/")[2]
-
-                    # Normalize the URL
-                    try:
-                        normalized_href = normalize_url(href, url)
-                    except ValueError:
-                        # logging.warning(f"Invalid URL format: {href}, Error: {str(e)}")
-                        return False
-
-                    link_data = {
-                        "href": normalized_href,
-                        "text": element.get_text().strip(),
-                        "title": element.get("title", "").strip(),
-                        "base_domain": base_domain,
-                    }
-
-                    is_external = is_external_url(normalized_href, base_domain)
-
-                    keep_element = True
-
-                    # Handle external link exclusions
-                    if is_external:
-                        link_base_domain = get_base_domain(normalized_href)
-                        link_data["base_domain"] = link_base_domain
-                        if kwargs.get("exclude_external_links", False):
-                            element.decompose()
-                            return False
-                        # elif kwargs.get('exclude_social_media_links', False):
-                        #     if link_base_domain in exclude_social_media_domains:
-                        #         element.decompose()
-                        #         return False
-                        # if any(domain in normalized_href.lower() for domain in exclude_social_media_domains):
-                        #     element.decompose()
-                        #     return False
-                        elif exclude_domains:
-                            if link_base_domain in exclude_domains:
-                                element.decompose()
-                                return False
-                            # if any(domain in normalized_href.lower() for domain in kwargs.get('exclude_domains', [])):
-                            #     element.decompose()
-                            #     return False
-
-                    if is_external:
-                        if normalized_href not in external_links_dict:
-                            external_links_dict[normalized_href] = link_data
-                    else:
-                        if kwargs.get("exclude_internal_links", False):
-                            element.decompose()
-                            return False
-                        if normalized_href not in internal_links_dict:
-                            internal_links_dict[normalized_href] = link_data
-
-            except Exception as e:
-                raise Exception(f"Error processing links: {str(e)}")
-
-            try:
-                if element.name == "img":
-                    potential_sources = [
-                        "src",
-                        "data-src",
-                        "srcset" "data-lazy-src",
-                        "data-original",
-                    ]
-                    src = element.get("src", "")
-                    while not src and potential_sources:
-                        src = element.get(potential_sources.pop(0), "")
-                    if not src:
-                        element.decompose()
-                        return False
-
-                    # If it is srcset pick up the first image
-                    if "srcset" in element.attrs:
-                        src = element.attrs["srcset"].split(",")[0].split(" ")[0]
-
-                    # If image src is internal, then skip
-                    if not is_external_url(src, base_domain):
-                        return True
-
-                    image_src_base_domain = get_base_domain(src)
-
-                    # Check flag if we should remove external images
-                    if kwargs.get("exclude_external_images", False):
-                        # Handle relative URLs (which are always from the same domain)
-                        if not src.startswith('http') and not src.startswith('//'):
-                            return True  # Keep relative URLs
-                        
-                        # For absolute URLs, compare the base domains using the existing function
-                        src_base_domain = get_base_domain(src)
-                        url_base_domain = get_base_domain(url)
-                        
-                        # If the domains don't match and both are valid, the image is external
-                        if src_base_domain and url_base_domain and src_base_domain != url_base_domain:
-                            element.decompose()
-                            return False
-
-                    # if kwargs.get('exclude_social_media_links', False):
-                    #     if image_src_base_domain in exclude_social_media_domains:
-                    #         element.decompose()
-                    #         return False
-                    # src_url_base = src.split('/')[2]
-                    # url_base = url.split('/')[2]
-                    # if any(domain in src for domain in exclude_social_media_domains):
-                    #     element.decompose()
-                    #     return False
-
-                    # Handle exclude domains
-                    if exclude_domains:
-                        if image_src_base_domain in exclude_domains:
-                            element.decompose()
-                            return False
-                        # if any(domain in src for domain in kwargs.get('exclude_domains', [])):
-                        #     element.decompose()
-                        #     return False
-
-                    return True  # Always keep image elements
-            except Exception:
-                raise "Error processing images"
-
-            # Check if flag to remove all forms is set
-            if kwargs.get("remove_forms", False) and element.name == "form":
-                element.decompose()
-                return False
-
-            if element.name in ["video", "audio"]:
-                media[f"{element.name}s"].append(
-                    {
-                        "src": element.get("src"),
-                        "alt": element.get("alt"),
-                        "type": element.name,
-                        "description": self.find_closest_parent_with_useful_text(
-                            element, **kwargs
-                        ),
-                    }
-                )
-                source_tags = element.find_all("source")
-                for source_tag in source_tags:
-                    media[f"{element.name}s"].append(
-                        {
-                            "src": source_tag.get("src"),
-                            "alt": element.get("alt"),
-                            "type": element.name,
-                            "description": self.find_closest_parent_with_useful_text(
-                                element, **kwargs
-                            ),
-                        }
-                    )
-                return True  # Always keep video and audio elements
-
-            if element.name in ONLY_TEXT_ELIGIBLE_TAGS:
-                if kwargs.get("only_text", False):
-                    element.replace_with(element.get_text())
-
-            try:
-                self.remove_unwanted_attributes(
-                    element, IMPORTANT_ATTRS + kwargs.get("keep_attrs", []) , kwargs.get("keep_data_attributes", False)
-                )
-            except Exception as e:
-                # print('Error removing unwanted attributes:', str(e))
-                self._log(
-                    "error",
-                    message="Error removing unwanted attributes: {error}",
-                    tag="SCRAPE",
-                    params={"error": str(e)},
-                )
-            # Process children
-            for child in list(element.children):
-                if isinstance(child, NavigableString) and not isinstance(
-                    child, Comment
-                ):
-                    if len(child.strip()) > 0:
-                        keep_element = True
-                else:
-                    if self._process_element(
-                        url,
-                        child,
-                        media,
-                        internal_links_dict,
-                        external_links_dict,
-                        **kwargs,
-                    ):
-                        keep_element = True
-
-            # Check word count
-            word_count_threshold = kwargs.get(
-                "word_count_threshold", MIN_WORD_THRESHOLD
-            )
-            if not keep_element:
-                word_count = len(element.get_text(strip=True).split())
-                keep_element = word_count >= word_count_threshold
-
-            if not keep_element:
-                element.decompose()
-
-            return keep_element
-        except Exception as e:
-            # print('Error processing element:', str(e))
-            self._log(
-                "error",
-                message="Error processing element: {error}",
-                tag="SCRAPE",
-                params={"error": str(e)},
-            )
-            return False
-
-    def _scrap(
-        self,
-        url: str,
-        html: str,
-        word_count_threshold: int = MIN_WORD_THRESHOLD,
-        css_selector: str = None,
-        target_elements: List[str] = None,
-        **kwargs,
-    ) -> Dict[str, Any]:
-        """
-        Extract content from HTML using BeautifulSoup.
-
-        Args:
-            url (str): The URL of the page to scrape.
-            html (str): The HTML content of the page to scrape.
-            word_count_threshold (int): The minimum word count threshold for content extraction.
-            css_selector (str): The CSS selector to use for content extraction.
-            **kwargs: Additional keyword arguments.
-
-        Returns:
-            dict: A dictionary containing the extracted content.
-        """
-        success = True
-        if not html:
-            return None
-
-        parser_type = kwargs.get("parser", "lxml")
-        soup = BeautifulSoup(html, parser_type)
-        body = soup.body
-        if body is None:
-            raise Exception("'<body>' tag is not found in fetched html. Consider adding wait_for=\"css:body\" to wait for body tag to be loaded into DOM.")
-        base_domain = get_base_domain(url)
-        
-        # Early removal of all images if exclude_all_images is set
-        # This happens before any processing to minimize memory usage
-        if kwargs.get("exclude_all_images", False):
-            for img in body.find_all('img'):
-                img.decompose()
-
-        try:
-            meta = extract_metadata("", soup)
-        except Exception as e:
-            self._log(
-                "error",
-                message="Error extracting metadata: {error}",
-                tag="SCRAPE",
-                params={"error": str(e)},
-            )
-            meta = {}
-
-        # Handle tag-based removal first - faster than CSS selection
-        excluded_tags = set(kwargs.get("excluded_tags", []) or [])
-        if excluded_tags:
-            for element in body.find_all(lambda tag: tag.name in excluded_tags):
-                element.extract()
-
-        # Handle CSS selector-based removal
-        excluded_selector = kwargs.get("excluded_selector", "")
-        if excluded_selector:
-            is_single_selector = (
-                "," not in excluded_selector and " " not in excluded_selector
-            )
-            if is_single_selector:
-                while element := body.select_one(excluded_selector):
-                    element.extract()
-            else:
-                for element in body.select(excluded_selector):
-                    element.extract()
-
-        content_element = None
-        if target_elements:
-            try:
-                for_content_targeted_element = []
-                for target_element in target_elements:
-                    for_content_targeted_element.extend(body.select(target_element))
-                content_element = soup.new_tag("div")
-                for el in for_content_targeted_element:
-                    content_element.append(copy.deepcopy(el))
-            except Exception as e:
-                self._log("error", f"Error with target element detection: {str(e)}", "SCRAPE")
-                return None
-        else:
-            content_element = body     
-
-        kwargs["exclude_social_media_domains"] = set(
-            kwargs.get("exclude_social_media_domains", []) + SOCIAL_MEDIA_DOMAINS
-        )
-        kwargs["exclude_domains"] = set(kwargs.get("exclude_domains", []))
-        if kwargs.get("exclude_social_media_links", False):
-            kwargs["exclude_domains"] = kwargs["exclude_domains"].union(
-                kwargs["exclude_social_media_domains"]
-            )
-
-        result_obj = self.process_element(
-            url,
-            body,
-            word_count_threshold=word_count_threshold,
-            base_domain=base_domain,
-            **kwargs,
-        )
-
-        links = {"internal": [], "external": []}
-        media = result_obj["media"]
-        internal_links_dict = result_obj["internal_links_dict"]
-        external_links_dict = result_obj["external_links_dict"]
-
-        # Update the links dictionary with unique links
-        links["internal"] = list(internal_links_dict.values())
-        links["external"] = list(external_links_dict.values())
-        
-        # Extract head content for links if configured
-        link_preview_config = kwargs.get("link_preview_config")
-        if link_preview_config is not None:
-            try:
-                import asyncio
-                from .link_preview import LinkPreview
-                from .models import Links, Link
-                
-                verbose = link_preview_config.verbose
-                
-                if verbose:
-                    self._log("info", "Starting link head extraction for {internal} internal and {external} external links",
-                              params={"internal": len(links["internal"]), "external": len(links["external"])}, tag="LINK_EXTRACT")
-                
-                # Convert dict links to Link objects
-                internal_links = [Link(**link_data) for link_data in links["internal"]]
-                external_links = [Link(**link_data) for link_data in links["external"]]
-                links_obj = Links(internal=internal_links, external=external_links)
-                
-                # Create a config object for LinkPreview  
-                class TempCrawlerRunConfig:
-                    def __init__(self, link_config, score_links):
-                        self.link_preview_config = link_config
-                        self.score_links = score_links
-                
-                config = TempCrawlerRunConfig(link_preview_config, kwargs.get("score_links", False))
-                
-                # Extract head content (run async operation in sync context)
-                async def extract_links():
-                    async with LinkPreview(self.logger) as extractor:
-                        return await extractor.extract_link_heads(links_obj, config)
-                
-                # Run the async operation
-                try:
-                    # Check if we're already in an async context
-                    loop = asyncio.get_running_loop()
-                    # If we're in an async context, we need to run in a thread
-                    import concurrent.futures
-                    with concurrent.futures.ThreadPoolExecutor() as executor:
-                        future = executor.submit(asyncio.run, extract_links())
-                        updated_links = future.result()
-                except RuntimeError:
-                    # No running loop, we can use asyncio.run directly
-                    updated_links = asyncio.run(extract_links())
-                
-                # Convert back to dict format
-                links["internal"] = [link.dict() for link in updated_links.internal]
-                links["external"] = [link.dict() for link in updated_links.external]
-                
-                if verbose:
-                    successful_internal = len([l for l in updated_links.internal if l.head_extraction_status == "valid"])
-                    successful_external = len([l for l in updated_links.external if l.head_extraction_status == "valid"])
-                    self._log("info", "Link head extraction completed: {internal_success}/{internal_total} internal, {external_success}/{external_total} external",
-                              params={
-                                  "internal_success": successful_internal,
-                                  "internal_total": len(updated_links.internal),
-                                  "external_success": successful_external,
-                                  "external_total": len(updated_links.external)
-                              }, tag="LINK_EXTRACT")
-                else:
-                    self._log("info", "Link head extraction completed successfully", tag="LINK_EXTRACT")
-                
-            except Exception as e:
-                self._log("error", f"Link head extraction failed: {str(e)}", tag="LINK_EXTRACT")
-                # Continue with original links if extraction fails
-
-        # # Process images using ThreadPoolExecutor
-        imgs = body.find_all("img")
-
-        media["images"] = [
-            img
-            for result in (
-                self.process_image(img, url, i, len(imgs), **kwargs)
-                for i, img in enumerate(imgs)
-            )
-            if result is not None
-            for img in result
-        ]
-        
-        # Process tables if not excluded
-        excluded_tags = set(kwargs.get("excluded_tags", []) or [])
-        if 'table' not in excluded_tags:
-            tables = body.find_all('table')
-            for table in tables:
-                if self.is_data_table(table, **kwargs):
-                    table_data = self.extract_table_data(table)
-                    media["tables"].append(table_data)
-
-        body = self.flatten_nested_elements(body)
-        base64_pattern = re.compile(r'data:image/[^;]+;base64,([^"]+)')
-        for img in imgs:
-            src = img.get("src", "")
-            if base64_pattern.match(src):
-                # Replace base64 data with empty string
-                img["src"] = base64_pattern.sub("", src)
-
-        str_body = ""
-        try:
-            str_body = content_element.encode_contents().decode("utf-8")
-        except Exception:
-            # Reset body to the original HTML
-            success = False
-            body = BeautifulSoup(html, "html.parser")
-
-            # Create a new div with a special ID
-            error_div = body.new_tag("div", id="crawl4ai_error_message")
-            error_div.string = """
-            Crawl4AI Error: This page is not fully supported.
-            
-            Possible reasons:
-            1. The page may have restrictions that prevent crawling.
-            2. The page might not be fully loaded.
-            
-            Suggestions:
-            - Try calling the crawl function with these parameters:
-            magic=True,
-            - Set headless=False to visualize what's happening on the page.
-            
-            If the issue persists, please check the page's structure and any potential anti-crawling measures.
-            """
-
-            # Append the error div to the body
-            body.append(error_div)
-            str_body = body.encode_contents().decode("utf-8")
-
-            print(
-                "[LOG] 😧 Error: After processing the crawled HTML and removing irrelevant tags, nothing was left in the page. Check the markdown for further details."
-            )
-            self._log(
-                "error",
-                message="After processing the crawled HTML and removing irrelevant tags, nothing was left in the page. Check the markdown for further details.",
-                tag="SCRAPE",
-            )
-
-        cleaned_html = str_body.replace("\n\n", "\n").replace("  ", " ")
-
-        return {
-            "cleaned_html": cleaned_html,
-            "success": success,
-            "media": media,
-            "links": links,
-            "metadata": meta,
-        }
-
-
-class LXMLWebScrapingStrategy(WebScrapingStrategy):
-    def __init__(self, logger=None):
-        super().__init__(logger)
-        self.DIMENSION_REGEX = re.compile(r"(\d+)(\D*)")
-        self.BASE64_PATTERN = re.compile(r'data:image/[^;]+;base64,([^"]+)')
-
     def _process_element(
         self,
         url: str,
@@ -1117,6 +242,16 @@ class LXMLWebScrapingStrategy(WebScrapingStrategy):
         exclude_domains = set(kwargs.get("exclude_domains", []))
 
         # Process links
+        try:
+            base_element = element.xpath("//head/base[@href]")
+            if base_element:
+                base_href = base_element[0].get("href", "").strip()
+                if base_href:
+                    url = base_href
+        except Exception as e:
+            self._log("error", f"Error extracting base URL: {str(e)}", "SCRAPE")
+            pass
+
         for link in element.xpath(".//a[@href]"):
             href = link.get("href", "").strip()
             if not href:
@@ -1862,3 +997,7 @@ class LXMLWebScrapingStrategy(WebScrapingStrategy):
                 "links": {"internal": [], "external": []},
                 "metadata": {},
             }
+
+
+# Backward compatibility alias
+WebScrapingStrategy = LXMLWebScrapingStrategy

crawl4ai/deep_crawling/bfs_strategy.py

@@ -38,7 +38,14 @@ class BFSDeepCrawlStrategy(DeepCrawlStrategy):
         self.include_external = include_external
         self.score_threshold = score_threshold
         self.max_pages = max_pages
-        self.logger = logger or logging.getLogger(__name__)
+        # Type check for logger
+        if isinstance(logger, dict):
+            logging.getLogger(__name__).warning(
+                "BFSDeepCrawlStrategy received a dict as logger; falling back to default logger."
+            )
+            self.logger = logging.getLogger(__name__)
+        else:
+            self.logger = logger or logging.getLogger(__name__)
         self.stats = TraversalStats(start_time=datetime.now())
         self._cancel_event = asyncio.Event()
         self._pages_crawled = 0

crawl4ai/docker_client.py

@@ -30,7 +30,7 @@ class Crawl4aiDockerClient:
     def __init__(
         self,
         base_url: str = "http://localhost:8000",
-        timeout: float = 30.0,
+        timeout: float = 600.0,  # Increased to 10 minutes for crawling operations
         verify_ssl: bool = True,
         verbose: bool = True,
         log_file: Optional[str] = None
@@ -113,21 +113,12 @@ class Crawl4aiDockerClient:
         self.logger.info(f"Crawling {len(urls)} URLs {'(streaming)' if is_streaming else ''}", tag="CRAWL")
         
         if is_streaming:
-            async def stream_results() -> AsyncGenerator[CrawlResult, None]:
-                async with self._http_client.stream("POST", f"{self.base_url}/crawl/stream", json=data) as response:
-                    response.raise_for_status()
-                    async for line in response.aiter_lines():
-                        if line.strip():
-                            result = json.loads(line)
-                            if "error" in result:
-                                self.logger.error_status(url=result.get("url", "unknown"), error=result["error"])
-                                continue
-                            self.logger.url_status(url=result.get("url", "unknown"), success=True, timing=result.get("timing", 0.0))
-                            if result.get("status") == "completed":
-                                continue
-                            else:
-                                yield CrawlResult(**result)
-            return stream_results()
+            # For streaming, we need to return the async generator properly
+            # The caller should be able to do: async for result in await client.crawl(...)
+            async def streaming_wrapper():
+                async for result in self._stream_crawl_results(data):
+                    yield result
+            return streaming_wrapper()
         
         response = await self._request("POST", "/crawl", json=data)
         result_data = response.json()
@@ -138,6 +129,35 @@ class Crawl4aiDockerClient:
         self.logger.success(f"Crawl completed with {len(results)} results", tag="CRAWL")
         return results[0] if len(results) == 1 else results
 
+    async def _stream_crawl_results(self, data: Dict[str, Any]) -> AsyncGenerator[CrawlResult, None]:
+        """Internal method to handle streaming crawl results."""
+        async with self._http_client.stream("POST", f"{self.base_url}/crawl/stream", json=data) as response:
+            response.raise_for_status()
+            async for line in response.aiter_lines():
+                if line.strip():
+                    try:
+                        result = json.loads(line)
+                        if "error" in result:
+                            self.logger.error_status(url=result.get("url", "unknown"), error=result["error"])
+                            continue
+                        
+                        # Check if this is a crawl result (has required fields)
+                        if "url" in result and "success" in result:
+                            self.logger.url_status(url=result.get("url", "unknown"), success=result.get("success", False), timing=result.get("timing", 0.0))
+                            
+                            # Create CrawlResult object properly
+                            crawl_result = CrawlResult(**result)
+                            yield crawl_result
+                        # Skip status-only messages
+                        elif result.get("status") == "completed":
+                            continue
+                    except json.JSONDecodeError as e:
+                        self.logger.error(f"Failed to parse streaming response: {e}", tag="STREAM")
+                        continue
+                    except Exception as e:
+                        self.logger.error(f"Error processing streaming result: {e}", tag="STREAM")
+                        continue
+
     async def get_schema(self) -> Dict[str, Any]:
         """Retrieve configuration schemas."""
         response = await self._request("GET", "/schema")

crawl4ai/install.py

@@ -119,6 +119,32 @@ def install_playwright():
         logger.warning(
             f"Please run '{sys.executable} -m playwright install --with-deps' manually after the installation."
         )
+    
+    # Install Patchright browsers for undetected browser support
+    logger.info("Installing Patchright browsers for undetected mode...", tag="INIT")
+    try:
+        subprocess.check_call(
+            [
+                sys.executable,
+                "-m",
+                "patchright",
+                "install",
+                "--with-deps",
+                "--force",
+                "chromium",
+            ]
+        )
+        logger.success(
+            "Patchright installation completed successfully.", tag="COMPLETE"
+        )
+    except subprocess.CalledProcessError:
+        logger.warning(
+            f"Please run '{sys.executable} -m patchright install --with-deps' manually after the installation."
+        )
+    except Exception:
+        logger.warning(
+            f"Please run '{sys.executable} -m patchright install --with-deps' manually after the installation."
+        )
 
 
 def run_migration():

crawl4ai/legacy/web_crawler.py

@@ -11,7 +11,7 @@ from .extraction_strategy import *
 from .crawler_strategy import *
 from typing import List
 from concurrent.futures import ThreadPoolExecutor
-from .content_scraping_strategy import WebScrapingStrategy
+from ..content_scraping_strategy import LXMLWebScrapingStrategy as WebScrapingStrategy
 from .config import *
 import warnings
 import json

crawl4ai/memory_utils.py

@@ -0,0 +1,79 @@
+import psutil
+import platform
+import subprocess
+from typing import Tuple
+
+
+def get_true_available_memory_gb() -> float:
+    """Get truly available memory including inactive pages (cross-platform)"""
+    vm = psutil.virtual_memory()
+
+    if platform.system() == 'Darwin':  # macOS
+        # On macOS, we need to include inactive memory too
+        try:
+            # Use vm_stat to get accurate values
+            result = subprocess.run(['vm_stat'], capture_output=True, text=True)
+            lines = result.stdout.split('\n')
+
+            page_size = 16384  # macOS page size
+            pages = {}
+
+            for line in lines:
+                if 'Pages free:' in line:
+                    pages['free'] = int(line.split()[-1].rstrip('.'))
+                elif 'Pages inactive:' in line:
+                    pages['inactive'] = int(line.split()[-1].rstrip('.'))
+                elif 'Pages speculative:' in line:
+                    pages['speculative'] = int(line.split()[-1].rstrip('.'))
+                elif 'Pages purgeable:' in line:
+                    pages['purgeable'] = int(line.split()[-1].rstrip('.'))
+
+            # Calculate total available (free + inactive + speculative + purgeable)
+            total_available_pages = (
+                pages.get('free', 0) + 
+                pages.get('inactive', 0) + 
+                pages.get('speculative', 0) + 
+                pages.get('purgeable', 0)
+            )
+            available_gb = (total_available_pages * page_size) / (1024**3)
+
+            return available_gb
+        except:
+            # Fallback to psutil
+            return vm.available / (1024**3)
+    else:
+        # For Windows and Linux, psutil.available is accurate
+        return vm.available / (1024**3)
+
+
+def get_true_memory_usage_percent() -> float:
+    """
+    Get memory usage percentage that accounts for platform differences.
+    
+    Returns:
+        float: Memory usage percentage (0-100)
+    """
+    vm = psutil.virtual_memory()
+    total_gb = vm.total / (1024**3)
+    available_gb = get_true_available_memory_gb()
+    
+    # Calculate used percentage based on truly available memory
+    used_percent = 100.0 * (total_gb - available_gb) / total_gb
+    
+    # Ensure it's within valid range
+    return max(0.0, min(100.0, used_percent))
+
+
+def get_memory_stats() -> Tuple[float, float, float]:
+    """
+    Get comprehensive memory statistics.
+    
+    Returns:
+        Tuple[float, float, float]: (used_percent, available_gb, total_gb)
+    """
+    vm = psutil.virtual_memory()
+    total_gb = vm.total / (1024**3)
+    available_gb = get_true_available_memory_gb()
+    used_percent = get_true_memory_usage_percent()
+    
+    return used_percent, available_gb, total_gb

crawl4ai/models.py

@@ -1,4 +1,36 @@
-from pydantic import BaseModel, HttpUrl, PrivateAttr, Field
+
+"""
+Crawl4AI Models Module
+
+This module contains Pydantic models used throughout the Crawl4AI library.
+
+Key Features:
+- ORJSONModel: Base model with ORJSON serialization support
+- DeprecatedPropertiesMixin: Global system for handling deprecated properties
+- CrawlResult: Main result model with backward compatibility support
+
+Deprecated Properties System:
+The DeprecatedPropertiesMixin provides a global way to handle deprecated properties
+across all models. Instead of manually excluding deprecated properties in each
+model_dump() call, you can simply override the get_deprecated_properties() method:
+
+Example:
+    class MyModel(ORJSONModel):
+        name: str
+        old_field: Optional[str] = None
+        
+        def get_deprecated_properties(self) -> set[str]:
+            return {'old_field', 'another_deprecated_field'}
+        
+        @property
+        def old_field(self):
+            raise AttributeError("old_field is deprecated, use name instead")
+
+The system automatically excludes these properties from serialization, preventing
+property objects from appearing in JSON output.
+"""
+
+from pydantic import BaseModel, ConfigDict,HttpUrl, PrivateAttr, Field
 from typing import List, Dict, Optional, Callable, Awaitable, Union, Any
 from typing import AsyncGenerator
 from typing import Generic, TypeVar
@@ -8,7 +40,7 @@ from .ssl_certificate import SSLCertificate
 from datetime import datetime
 from datetime import timedelta
 
-
+import orjson
 ###############################
 # Dispatcher Models
 ###############################
@@ -91,7 +123,122 @@ class TokenUsage:
     completion_tokens_details: Optional[dict] = None
     prompt_tokens_details: Optional[dict] = None
 
-class UrlModel(BaseModel):
+
+def orjson_default(obj):
+    # Handle datetime (if not already handled by orjson)
+    if isinstance(obj, datetime):
+        return obj.isoformat()
+
+    # Handle property objects (convert to string or something else)
+    if isinstance(obj, property):
+        return str(obj)
+
+    # Last resort: convert to string
+    return str(obj)
+
+
+class DeprecatedPropertiesMixin:
+    """
+    Mixin to handle deprecated properties in Pydantic models.
+    
+    Classes that inherit from this mixin can define deprecated properties
+    that will be automatically excluded from serialization.
+    
+    Usage:
+    1. Override the get_deprecated_properties() method to return a set of deprecated property names
+    2. The model_dump method will automatically exclude these properties
+    
+    Example:
+        class MyModel(ORJSONModel):
+            def get_deprecated_properties(self) -> set[str]:
+                return {'old_field', 'legacy_property'}
+            
+            name: str
+            old_field: Optional[str] = None  # Field definition
+            
+            @property
+            def old_field(self):  # Property that overrides the field
+                raise AttributeError("old_field is deprecated, use name instead")
+    """
+    
+    def get_deprecated_properties(self) -> set[str]:
+        """
+        Get deprecated property names for this model.
+        Override this method in subclasses to define deprecated properties.
+        
+        Returns:
+            set[str]: Set of deprecated property names
+        """
+        return set()
+    
+    @classmethod
+    def get_all_deprecated_properties(cls) -> set[str]:
+        """
+        Get all deprecated properties from this class and all parent classes.
+        
+        Returns:
+            set[str]: Set of all deprecated property names
+        """
+        deprecated_props = set()
+        # Create an instance to call the instance method
+        try:
+            # Try to create a dummy instance to get deprecated properties
+            dummy_instance = cls.__new__(cls)
+            deprecated_props.update(dummy_instance.get_deprecated_properties())
+        except Exception:
+            # If we can't create an instance, check for class-level definitions
+            pass
+            
+        # Also check parent classes
+        for klass in cls.__mro__:
+            if hasattr(klass, 'get_deprecated_properties') and klass != DeprecatedPropertiesMixin:
+                try:
+                    dummy_instance = klass.__new__(klass)
+                    deprecated_props.update(dummy_instance.get_deprecated_properties())
+                except Exception:
+                    pass
+        return deprecated_props
+    
+    def model_dump(self, *args, **kwargs):
+        """
+        Override model_dump to automatically exclude deprecated properties.
+        
+        This method:
+        1. Gets the existing exclude set from kwargs
+        2. Adds all deprecated properties defined in get_deprecated_properties()
+        3. Calls the parent model_dump with the updated exclude set
+        """
+        # Get the default exclude set, or create empty set if None
+        exclude = kwargs.get('exclude', set())
+        if exclude is None:
+            exclude = set()
+        elif not isinstance(exclude, set):
+            exclude = set(exclude) if exclude else set()
+        
+        # Add deprecated properties for this instance
+        exclude.update(self.get_deprecated_properties())
+        kwargs['exclude'] = exclude
+        
+        return super().model_dump(*args, **kwargs)
+
+
+class ORJSONModel(DeprecatedPropertiesMixin, BaseModel):
+    model_config = ConfigDict(
+        ser_json_timedelta="iso8601",  # Optional: format timedelta
+        ser_json_bytes="utf8",         # Optional: bytes → UTF-8 string
+    )
+    
+    def model_dump_json(self, **kwargs) -> bytes:
+        """Custom JSON serialization using orjson"""
+        return orjson.dumps(self.model_dump(**kwargs), default=orjson_default)
+    
+    @classmethod
+    def model_validate_json(cls, json_data: Union[str, bytes], **kwargs):
+        """Custom JSON deserialization using orjson"""
+        if isinstance(json_data, str):
+            json_data = json_data.encode()
+        return cls.model_validate(orjson.loads(json_data), **kwargs)
+class UrlModel(ORJSONModel):
     url: HttpUrl
     forced: bool = False
 
@@ -108,7 +255,7 @@ class TraversalStats:
     total_depth_reached: int = 0
     current_depth: int = 0
 
-class DispatchResult(BaseModel):
+class DispatchResult(ORJSONModel):
     task_id: str
     memory_usage: float
     peak_memory: float
@@ -116,7 +263,7 @@ class DispatchResult(BaseModel):
     end_time: Union[datetime, float]
     error_message: str = ""
 
-class MarkdownGenerationResult(BaseModel):
+class MarkdownGenerationResult(ORJSONModel):
     raw_markdown: str
     markdown_with_citations: str
     references_markdown: str
@@ -126,7 +273,7 @@ class MarkdownGenerationResult(BaseModel):
     def __str__(self):
         return self.raw_markdown
     
-class CrawlResult(BaseModel):
+class CrawlResult(ORJSONModel):
     url: str
     html: str
     fit_html: Optional[str] = None
@@ -156,6 +303,10 @@ class CrawlResult(BaseModel):
     class Config:
         arbitrary_types_allowed = True
 
+    def get_deprecated_properties(self) -> set[str]:
+        """Define deprecated properties that should be excluded from serialization."""
+        return {'fit_html', 'fit_markdown', 'markdown_v2'}
+
 # NOTE: The StringCompatibleMarkdown class, custom __init__ method, property getters/setters,
 # and model_dump override all exist to support a smooth transition from markdown as a string
 # to markdown as a MarkdownGenerationResult object, while maintaining backward compatibility.
@@ -245,14 +396,16 @@ class CrawlResult(BaseModel):
         1. PrivateAttr fields are excluded from serialization by default
         2. We need to maintain backward compatibility by including the 'markdown' field
            in the serialized output
-        3. We're transitioning from 'markdown_v2' to enhancing 'markdown' to hold
-           the same type of data
+        3. Uses the DeprecatedPropertiesMixin to automatically exclude deprecated properties
         
         Future developers: This method ensures that the markdown content is properly
-        serialized despite being stored in a private attribute. If the serialization
-        requirements change, this is where you would update the logic.
+        serialized despite being stored in a private attribute. The deprecated properties
+        are automatically handled by the mixin.
         """
+        # Use the parent class method which handles deprecated properties automatically
         result = super().model_dump(*args, **kwargs)
+        
+        # Add the markdown content if it exists
         if self._markdown is not None:
             result["markdown"] = self._markdown.model_dump() 
         return result
@@ -307,7 +460,7 @@ RunManyReturn = Union[
 # 1. Replace the private attribute and property with a standard field
 # 2. Update any serialization logic that might depend on the current behavior
 
-class AsyncCrawlResponse(BaseModel):
+class AsyncCrawlResponse(ORJSONModel):
     html: str
     response_headers: Dict[str, str]
     js_execution_result: Optional[Dict[str, Any]] = None
@@ -328,7 +481,7 @@ class AsyncCrawlResponse(BaseModel):
 ###############################
 # Scraping Models
 ###############################
-class MediaItem(BaseModel):
+class MediaItem(ORJSONModel):
     src: Optional[str] = ""
     data: Optional[str] = ""
     alt: Optional[str] = ""
@@ -340,7 +493,7 @@ class MediaItem(BaseModel):
     width: Optional[int] = None
 
 
-class Link(BaseModel):
+class Link(ORJSONModel):
     href: Optional[str] = ""
     text: Optional[str] = ""
     title: Optional[str] = ""
@@ -353,7 +506,7 @@ class Link(BaseModel):
     total_score: Optional[float] = None  # Combined score from intrinsic and contextual scores
 
 
-class Media(BaseModel):
+class Media(ORJSONModel):
     images: List[MediaItem] = []
     videos: List[
         MediaItem
@@ -364,12 +517,12 @@ class Media(BaseModel):
     tables: List[Dict] = []  # Table data extracted from HTML tables
 
 
-class Links(BaseModel):
+class Links(ORJSONModel):
     internal: List[Link] = []
     external: List[Link] = []
 
 
-class ScrapingResult(BaseModel):
+class ScrapingResult(ORJSONModel):
     cleaned_html: str
     success: bool
     media: Media = Media()

crawl4ai/prompts.py

@@ -1056,7 +1056,7 @@ Your output must:
 </output_requirements>
 """
 
-GENERATE_SCRIPT_PROMPT = """You are a world-class browser automation specialist. Your sole purpose is to convert a natural language objective and a snippet of HTML into the most **efficient, robust, and simple** script possible to prepare a web page for data extraction.
+GENERATE_SCRIPT_PROMPT = r"""You are a world-class browser automation specialist. Your sole purpose is to convert a natural language objective and a snippet of HTML into the most **efficient, robust, and simple** script possible to prepare a web page for data extraction.
 
 Your scripts run **before the crawl** to handle dynamic content, user interactions, and other obstacles. You are a master of two tools: raw **JavaScript** and the high-level **Crawl4ai Script (c4a)**.
 

crawl4ai/types.py

@@ -23,8 +23,9 @@ SeedingConfig = Union['SeedingConfigType']
 
 # Content scraping types
 ContentScrapingStrategy = Union['ContentScrapingStrategyType']
-WebScrapingStrategy = Union['WebScrapingStrategyType']
 LXMLWebScrapingStrategy = Union['LXMLWebScrapingStrategyType']
+# Backward compatibility alias
+WebScrapingStrategy = Union['LXMLWebScrapingStrategyType']
 
 # Proxy types
 ProxyRotationStrategy = Union['ProxyRotationStrategyType']
@@ -114,7 +115,6 @@ if TYPE_CHECKING:
     # Content scraping imports
     from .content_scraping_strategy import (
         ContentScrapingStrategy as ContentScrapingStrategyType,
-        WebScrapingStrategy as WebScrapingStrategyType,
         LXMLWebScrapingStrategy as LXMLWebScrapingStrategyType,
     )
     

crawl4ai/utils.py

@@ -1517,8 +1517,29 @@ def extract_metadata_using_lxml(html, doc=None):
     head = head[0]
 
     # Title - using XPath
+    # title = head.xpath(".//title/text()")
+    # metadata["title"] = title[0].strip() if title else None
+
+    # === Title Extraction - New Approach ===
+    # Attempt to extract <title> using XPath
     title = head.xpath(".//title/text()")
-    metadata["title"] = title[0].strip() if title else None
+    title = title[0] if title else None
+
+    # Fallback: Use .find() in case XPath fails due to malformed HTML
+    if not title:
+        title_el = doc.find(".//title")
+        title = title_el.text if title_el is not None else None
+
+    # Final fallback: Use OpenGraph or Twitter title if <title> is missing or empty
+    if not title:
+        title_candidates = (
+            doc.xpath("//meta[@property='og:title']/@content") or
+            doc.xpath("//meta[@name='twitter:title']/@content")
+        )
+        title = title_candidates[0] if title_candidates else None
+
+    # Strip and assign title
+    metadata["title"] = title.strip() if title else None
 
     # Meta description - using XPath with multiple attribute conditions
     description = head.xpath('.//meta[@name="description"]/@content')
@@ -3342,7 +3363,13 @@ async def get_text_embeddings(
     # Default: use sentence-transformers
     else:
         # Lazy load to avoid importing heavy libraries unless needed
-        from sentence_transformers import SentenceTransformer
+        try:
+            from sentence_transformers import SentenceTransformer
+        except ImportError:
+            raise ImportError(
+                "sentence-transformers is required for local embeddings. "
+                "Install it with: pip install 'crawl4ai[transformer]' or pip install sentence-transformers"
+            )
         
         # Cache the model in function attribute to avoid reloading
         if not hasattr(get_text_embeddings, '_models'):

deploy/docker/.llm.env.example

@@ -5,4 +5,9 @@ ANTHROPIC_API_KEY=your_anthropic_key_here
 GROQ_API_KEY=your_groq_key_here
 TOGETHER_API_KEY=your_together_key_here
 MISTRAL_API_KEY=your_mistral_key_here
-GEMINI_API_TOKEN=your_gemini_key_here
+GEMINI_API_TOKEN=your_gemini_key_here
+
+# Optional: Override the default LLM provider
+# Examples: "openai/gpt-4", "anthropic/claude-3-opus", "deepseek/chat", etc.
+# If not set, uses the provider specified in config.yml (default: openai/gpt-4o-mini)
+# LLM_PROVIDER=anthropic/claude-3-opus

deploy/docker/README.md

@@ -154,6 +154,29 @@ cp deploy/docker/.llm.env.example .llm.env
 # Now edit .llm.env and add your API keys
 ```
 
+**Flexible LLM Provider Configuration:**
+
+The Docker setup now supports flexible LLM provider configuration through three methods:
+
+1. **Environment Variable** (Highest Priority): Set `LLM_PROVIDER` to override the default
+   ```bash
+   export LLM_PROVIDER="anthropic/claude-3-opus"
+   # Or in your .llm.env file:
+   # LLM_PROVIDER=anthropic/claude-3-opus
+   ```
+
+2. **API Request Parameter**: Specify provider per request
+   ```json
+   {
+     "url": "https://example.com",
+     "provider": "groq/mixtral-8x7b"
+   }
+   ```
+
+3. **Config File Default**: Falls back to `config.yml` (default: `openai/gpt-4o-mini`)
+
+The system automatically selects the appropriate API key based on the provider.
+
 #### 3. Build and Run with Compose
 
 The `docker-compose.yml` file in the project root provides a simplified approach that automatically handles architecture detection using buildx.
@@ -668,7 +691,7 @@ app:
 
 # Default LLM Configuration
 llm:
-  provider: "openai/gpt-4o-mini"
+  provider: "openai/gpt-4o-mini"  # Can be overridden by LLM_PROVIDER env var
   api_key_env: "OPENAI_API_KEY"
   # api_key: sk-...  # If you pass the API key directly then api_key_env will be ignored
 

deploy/docker/api.py

@@ -1,10 +1,12 @@
 import os
 import json
+import orjson
 import asyncio
 from typing import List, Tuple, Dict
 from functools import partial
 from uuid import uuid4
 from datetime import datetime
+from base64 import b64encode
 
 import logging
 from typing import Optional, AsyncGenerator
@@ -39,7 +41,9 @@ from utils import (
     get_base_url,
     is_task_id,
     should_cleanup_task,
-    decode_redis_hash
+    decode_redis_hash,
+    get_llm_api_key,
+    validate_llm_provider
 )
 
 import psutil, time
@@ -62,7 +66,7 @@ async def handle_llm_qa(
 ) -> str:
     """Process QA using LLM with crawled content as context."""
     try:
-        if not url.startswith(('http://', 'https://')):
+        if not url.startswith(('http://', 'https://')) and not url.startswith(("raw:", "raw://")):
             url = 'https://' + url
         # Extract base URL by finding last '?q=' occurrence
         last_q_index = url.rfind('?q=')
@@ -88,10 +92,12 @@ async def handle_llm_qa(
 
     Answer:"""
 
+        # api_token=os.environ.get(config["llm"].get("api_key_env", ""))
+
         response = perform_completion_with_backoff(
             provider=config["llm"]["provider"],
             prompt_with_variables=prompt,
-            api_token=os.environ.get(config["llm"].get("api_key_env", ""))
+            api_token=get_llm_api_key(config)
         )
 
         return response.choices[0].message.content
@@ -109,19 +115,23 @@ async def process_llm_extraction(
     url: str,
     instruction: str,
     schema: Optional[str] = None,
-    cache: str = "0"
+    cache: str = "0",
+    provider: Optional[str] = None
 ) -> None:
     """Process LLM extraction in background."""
     try:
-        # If config['llm'] has api_key then ignore the api_key_env
-        api_key = ""
-        if "api_key" in config["llm"]:
-            api_key = config["llm"]["api_key"]
-        else:
-            api_key = os.environ.get(config["llm"].get("api_key_env", None), "")
+        # Validate provider
+        is_valid, error_msg = validate_llm_provider(config, provider)
+        if not is_valid:
+            await redis.hset(f"task:{task_id}", mapping={
+                "status": TaskStatus.FAILED,
+                "error": error_msg
+            })
+            return
+        api_key = get_llm_api_key(config, provider)
         llm_strategy = LLMExtractionStrategy(
             llm_config=LLMConfig(
-                provider=config["llm"]["provider"],
+                provider=provider or config["llm"]["provider"],
                 api_token=api_key
             ),
             instruction=instruction,
@@ -168,12 +178,21 @@ async def handle_markdown_request(
     filter_type: FilterType,
     query: Optional[str] = None,
     cache: str = "0",
-    config: Optional[dict] = None
+    config: Optional[dict] = None,
+    provider: Optional[str] = None
 ) -> str:
     """Handle markdown generation requests."""
     try:
+        # Validate provider if using LLM filter
+        if filter_type == FilterType.LLM:
+            is_valid, error_msg = validate_llm_provider(config, provider)
+            if not is_valid:
+                raise HTTPException(
+                    status_code=status.HTTP_400_BAD_REQUEST,
+                    detail=error_msg
+                )
         decoded_url = unquote(url)
-        if not decoded_url.startswith(('http://', 'https://')):
+        if not decoded_url.startswith(('http://', 'https://')) and not decoded_url.startswith(("raw:", "raw://")):
             decoded_url = 'https://' + decoded_url
 
         if filter_type == FilterType.RAW:
@@ -184,8 +203,8 @@ async def handle_markdown_request(
                 FilterType.BM25: BM25ContentFilter(user_query=query or ""),
                 FilterType.LLM: LLMContentFilter(
                     llm_config=LLMConfig(
-                        provider=config["llm"]["provider"],
-                        api_token=os.environ.get(config["llm"].get("api_key_env", None), ""),
+                        provider=provider or config["llm"]["provider"],
+                        api_token=get_llm_api_key(config, provider),
                     ),
                     instruction=query or "Extract main content"
                 )
@@ -229,7 +248,8 @@ async def handle_llm_request(
     query: Optional[str] = None,
     schema: Optional[str] = None,
     cache: str = "0",
-    config: Optional[dict] = None
+    config: Optional[dict] = None,
+    provider: Optional[str] = None
 ) -> JSONResponse:
     """Handle LLM extraction requests."""
     base_url = get_base_url(request)
@@ -259,7 +279,8 @@ async def handle_llm_request(
             schema,
             cache,
             base_url,
-            config
+            config,
+            provider
         )
 
     except Exception as e:
@@ -303,11 +324,12 @@ async def create_new_task(
     schema: Optional[str],
     cache: str,
     base_url: str,
-    config: dict
+    config: dict,
+    provider: Optional[str] = None
 ) -> JSONResponse:
     """Create and initialize a new task."""
     decoded_url = unquote(input_path)
-    if not decoded_url.startswith(('http://', 'https://')):
+    if not decoded_url.startswith(('http://', 'https://')) and not decoded_url.startswith(("raw:", "raw://")):
         decoded_url = 'https://' + decoded_url
 
     from datetime import datetime
@@ -327,7 +349,8 @@ async def create_new_task(
         decoded_url,
         query,
         schema,
-        cache
+        cache,
+        provider
     )
 
     return JSONResponse({
@@ -362,24 +385,60 @@ def create_task_response(task: dict, task_id: str, base_url: str) -> dict:
 
 async def stream_results(crawler: AsyncWebCrawler, results_gen: AsyncGenerator) -> AsyncGenerator[bytes, None]:
     """Stream results with heartbeats and completion markers."""
-    import json
-    from utils import datetime_handler
+    import orjson
+    from datetime import datetime
+    import inspect
+    
+    def orjson_default(obj):
+        # Handle datetime (if not already handled by orjson)
+        if isinstance(obj, datetime):
+            return obj.isoformat()
+        # Handle property objects (convert to string or something else)
+        if isinstance(obj, property):
+            return str(obj)
+        # Last resort: convert to string
+        return str(obj)
 
     try:
-        async for result in results_gen:
-            try:
-                server_memory_mb = _get_memory_mb()
-                result_dict = result.model_dump()
-                result_dict['server_memory_mb'] = server_memory_mb
-                logger.info(f"Streaming result for {result_dict.get('url', 'unknown')}")
-                data = json.dumps(result_dict, default=datetime_handler) + "\n"
-                yield data.encode('utf-8')
-            except Exception as e:
-                logger.error(f"Serialization error: {e}")
-                error_response = {"error": str(e), "url": getattr(result, 'url', 'unknown')}
-                yield (json.dumps(error_response) + "\n").encode('utf-8')
+        logger.info(f"Starting streaming with results_gen type: {type(results_gen)}")
+        logger.info(f"Is results_gen async generator: {inspect.isasyncgen(results_gen)}")
+        
+        # Check if results_gen is actually an async generator vs another type
+        if inspect.isasyncgen(results_gen):
+            logger.info("Processing as async generator")
+            async for result in results_gen:
+                try:
+                    logger.info(f"Processing streaming result of type: {type(result)}")
+                    
+                    # Check if this result is actually a CrawlResult
+                    if hasattr(result, 'model_dump_json'):
+                        server_memory_mb = _get_memory_mb()
+                        result_json = result.model_dump_json()
+                        result_dict = orjson.loads(result_json)
+                        result_dict['server_memory_mb'] = server_memory_mb
+                        
+                        if result_dict.get('pdf') is not None:
+                            result_dict['pdf'] = b64encode(result_dict['pdf']).decode('utf-8')
+                        
+                        logger.info(f"Streaming result for {result_dict.get('url', 'unknown')}")
+                        data = orjson.dumps(result_dict, default=orjson_default).decode('utf-8') + "\n"
+                        yield data.encode('utf-8')
+                    else:
+                        logger.error(f"Result doesn't have model_dump_json method: {type(result)}")
+                        error_response = {"error": f"Invalid result type: {type(result)}", "url": "unknown"}
+                        yield (orjson.dumps(error_response).decode('utf-8') + "\n").encode('utf-8')
+                        
+                except Exception as e:
+                    logger.error(f"Serialization error: {e}")
+                    logger.error(f"Result type was: {type(result)}")
+                    error_response = {"error": str(e), "url": getattr(result, 'url', 'unknown')}
+                    yield (orjson.dumps(error_response).decode('utf-8') + "\n").encode('utf-8')
+        else:
+            logger.error(f"results_gen is not an async generator: {type(results_gen)}")
+            error_response = {"error": f"Invalid results_gen type: {type(results_gen)}"}
+            yield (orjson.dumps(error_response).decode('utf-8') + "\n").encode('utf-8')
 
-        yield json.dumps({"status": "completed"}).encode('utf-8')
+        yield orjson.dumps({"status": "completed"}).decode('utf-8').encode('utf-8')
         
     except asyncio.CancelledError:
         logger.warning("Client disconnected during streaming")
@@ -403,7 +462,7 @@ async def handle_crawl_request(
     peak_mem_mb = start_mem_mb
     
     try:
-        urls = [('https://' + url) if not url.startswith(('http://', 'https://')) else url for url in urls]
+        urls = [('https://' + url) if not url.startswith(('http://', 'https://')) and not url.startswith(("raw:", "raw://")) else url for url in urls]
         browser_config = BrowserConfig.load(browser_config)
         crawler_config = CrawlerRunConfig.load(crawler_config)
 
@@ -443,10 +502,21 @@ async def handle_crawl_request(
             mem_delta_mb = end_mem_mb - start_mem_mb # <--- Calculate delta
             peak_mem_mb = max(peak_mem_mb if peak_mem_mb else 0, end_mem_mb) # <--- Get peak memory
         logger.info(f"Memory usage: Start: {start_mem_mb} MB, End: {end_mem_mb} MB, Delta: {mem_delta_mb} MB, Peak: {peak_mem_mb} MB")
-                              
+
+        # Process results to handle PDF bytes
+        processed_results = []
+        for result in results:
+            # Use ORJSON serialization to handle property objects properly
+            result_json = result.model_dump_json()
+            result_dict = orjson.loads(result_json)
+            # If PDF exists, encode it to base64
+            if result_dict.get('pdf') is not None:
+                result_dict['pdf'] = b64encode(result_dict['pdf']).decode('utf-8')
+            processed_results.append(result_dict)
+            
         return {
             "success": True,
-            "results": [result.model_dump() for result in results],
+            "results": processed_results,
             "server_processing_time_s": end_time - start_time,
             "server_memory_delta_mb": mem_delta_mb,
             "server_peak_memory_mb": peak_mem_mb
@@ -488,8 +558,19 @@ async def handle_stream_crawl_request(
         browser_config.verbose = False
         crawler_config = CrawlerRunConfig.load(crawler_config)
         crawler_config.scraping_strategy = LXMLWebScrapingStrategy()
-        crawler_config.stream = True
+        # Don't force stream=True here - let the deep crawl strategy control its own streaming behavior
 
+        # Apply global base config (this was missing!)
+        base_config = config["crawler"]["base_config"]
+        for key, value in base_config.items():
+            if hasattr(crawler_config, key):
+                print(f"[DEBUG] Applying base_config: {key} = {value}")
+                setattr(crawler_config, key, value)
+
+        print(f"[DEBUG] Deep crawl strategy: {type(crawler_config.deep_crawl_strategy).__name__ if crawler_config.deep_crawl_strategy else 'None'}")
+        print(f"[DEBUG] Stream mode: {crawler_config.stream}")
+        print(f"[DEBUG] Simulate user: {getattr(crawler_config, 'simulate_user', 'Not set')}")
+        
         dispatcher = MemoryAdaptiveDispatcher(
             memory_threshold_percent=config["crawler"]["memory_threshold_percent"],
             rate_limiter=RateLimiter(
@@ -503,11 +584,58 @@ async def handle_stream_crawl_request(
         # crawler = AsyncWebCrawler(config=browser_config)
         # await crawler.start()
 
-        results_gen = await crawler.arun_many(
-            urls=urls,
-            config=crawler_config,
-            dispatcher=dispatcher
-        )
+        # Use correct method based on URL count (same as regular endpoint)
+        if len(urls) == 1:
+            # For single URL, use arun to get CrawlResult, then wrap in async generator
+            single_result_container = await crawler.arun(
+                url=urls[0],
+                config=crawler_config,
+                dispatcher=dispatcher
+            )
+            
+            async def single_result_generator():
+                # Handle CrawlResultContainer - extract the actual results
+                if hasattr(single_result_container, '_results'):
+                    # It's a CrawlResultContainer - iterate over the internal results
+                    for result in single_result_container._results:
+                        # Check if the result is an async generator (from deep crawl)
+                        if hasattr(result, '__aiter__'):
+                            async for sub_result in result:
+                                yield sub_result
+                        else:
+                            yield result
+                elif hasattr(single_result_container, '__aiter__'):
+                    # It's an async generator (from streaming deep crawl)
+                    async for result in single_result_container:
+                        yield result
+                elif hasattr(single_result_container, '__iter__') and not hasattr(single_result_container, 'url'):
+                    # It's iterable but not a CrawlResult itself
+                    for result in single_result_container:
+                        # Check if each result is an async generator
+                        if hasattr(result, '__aiter__'):
+                            async for sub_result in result:
+                                yield sub_result
+                        else:
+                            yield result
+                else:
+                    # It's a single CrawlResult
+                    yield single_result_container
+                    
+            results_gen = single_result_generator()
+        else:
+            # For multiple URLs, use arun_many
+            results_gen = await crawler.arun_many(
+                urls=urls,
+                config=crawler_config,
+                dispatcher=dispatcher
+            )
+            
+            # If results_gen is a list (e.g., from deep crawl), convert to async generator
+            if isinstance(results_gen, list):
+                async def convert_list_to_generator():
+                    for result in results_gen:
+                        yield result
+                results_gen = convert_list_to_generator()
 
         return crawler, results_gen
 

deploy/docker/job.py

@@ -36,6 +36,7 @@ class LlmJobPayload(BaseModel):
     q:      str
     schema: Optional[str] = None
     cache:  bool = False
+    provider: Optional[str] = None
 
 
 class CrawlJobPayload(BaseModel):
@@ -61,6 +62,7 @@ async def llm_job_enqueue(
         schema=payload.schema,
         cache=payload.cache,
         config=_config,
+        provider=payload.provider,
     )
 
 

deploy/docker/schemas.py

@@ -15,6 +15,7 @@ class MarkdownRequest(BaseModel):
     f:   FilterType             = Field(FilterType.FIT, description="Content‑filter strategy: fit, raw, bm25, or llm")
     q:   Optional[str] = Field(None,  description="Query string used by BM25/LLM filters")
     c:   Optional[str] = Field("0",   description="Cache‑bust / revision counter")
+    provider: Optional[str] = Field(None, description="LLM provider override (e.g., 'anthropic/claude-3-opus')")
 
 
 class RawCode(BaseModel):

deploy/docker/server.py

@@ -7,13 +7,16 @@ Crawl4AI FastAPI entry‑point
 """
 
 # ── stdlib & 3rd‑party imports ───────────────────────────────
+from datetime import datetime
+
+import orjson
 from crawler_pool import get_crawler, close_all, janitor
 from crawl4ai import AsyncWebCrawler, BrowserConfig, CrawlerRunConfig
 from auth import create_access_token, get_token_dependency, TokenRequest
 from pydantic import BaseModel
 from typing import Optional, List, Dict
 from fastapi import Request, Depends
-from fastapi.responses import FileResponse
+from fastapi.responses import FileResponse, ORJSONResponse
 import base64
 import re
 from crawl4ai import AsyncWebCrawler, BrowserConfig, CrawlerRunConfig
@@ -32,6 +35,8 @@ from schemas import (
     JSEndpointRequest,
 )
 
+# Use the proper serialization functions from async_configs
+from crawl4ai.async_configs import to_serializable_dict
 from utils import (
     FilterType, load_config, setup_logging, verify_email_domain
 )
@@ -112,11 +117,26 @@ async def lifespan(_: FastAPI):
     app.state.janitor.cancel()
     await close_all()
 
+def orjson_default(obj):
+    # Handle datetime (if not already handled by orjson)
+    if isinstance(obj, datetime):
+        return obj.isoformat()
+
+    # Handle property objects (convert to string or something else)
+    if isinstance(obj, property):
+        return str(obj)
+
+    # Last resort: convert to string
+    return str(obj)
+
+def orjson_dumps(v, *, default):
+    return orjson.dumps(v, default=orjson_default).decode()
 # ───────────────────── FastAPI instance ──────────────────────
 app = FastAPI(
     title=config["app"]["title"],
     version=config["app"]["version"],
     lifespan=lifespan,
+    default_response_class=ORJSONResponse
 )
 
 # ── static playground ──────────────────────────────────────
@@ -237,11 +257,11 @@ async def get_markdown(
     body: MarkdownRequest,
     _td: Dict = Depends(token_dep),
 ):
-    if not body.url.startswith(("http://", "https://")):
+    if not body.url.startswith(("http://", "https://")) and not body.url.startswith(("raw:", "raw://")):
         raise HTTPException(
-            400, "URL must be absolute and start with http/https")
+            400, "Invalid URL format. Must start with http://, https://, or for raw HTML (raw:, raw://)")
     markdown = await handle_markdown_request(
-        body.url, body.f, body.q, body.c, config
+        body.url, body.f, body.q, body.c, config, body.provider
     )
     return JSONResponse({
         "url": body.url,
@@ -401,7 +421,7 @@ async def llm_endpoint(
 ):
     if not q:
         raise HTTPException(400, "Query parameter 'q' is required")
-    if not url.startswith(("http://", "https://")):
+    if not url.startswith(("http://", "https://")) and not url.startswith(("raw:", "raw://")):
         url = "https://" + url
     answer = await handle_llm_qa(url, q, config)
     return JSONResponse({"answer": answer})
@@ -435,15 +455,20 @@ async def crawl(
     """
     Crawl a list of URLs and return the results as JSON.
     """
-    if not crawl_request.urls:
-        raise HTTPException(400, "At least one URL required")
-    res = await handle_crawl_request(
-        urls=crawl_request.urls,
-        browser_config=crawl_request.browser_config,
-        crawler_config=crawl_request.crawler_config,
-        config=config,
-    )
-    return JSONResponse(res)
+    try:
+        if not crawl_request.urls:
+            raise HTTPException(400, "At least one URL required")
+        res = await handle_crawl_request(
+            urls=crawl_request.urls,
+            browser_config=crawl_request.browser_config,
+            crawler_config=crawl_request.crawler_config,
+            config=config,
+        )
+        # handle_crawl_request returns a dictionary, so we can pass it directly to ORJSONResponse
+        return ORJSONResponse(res)
+    except Exception as e:
+        print(f"Error occurred: {e}")
+        return ORJSONResponse({"error": str(e)}, status_code=500)
 
 
 @app.post("/crawl/stream")

deploy/docker/utils.py

@@ -1,6 +1,7 @@
 import dns.resolver
 import logging
 import yaml
+import os
 from datetime import datetime
 from enum import Enum
 from pathlib import Path
@@ -19,10 +20,24 @@ class FilterType(str, Enum):
     LLM = "llm"
 
 def load_config() -> Dict:
-    """Load and return application configuration."""
+    """Load and return application configuration with environment variable overrides."""
     config_path = Path(__file__).parent / "config.yml"
     with open(config_path, "r") as config_file:
-        return yaml.safe_load(config_file)
+        config = yaml.safe_load(config_file)
+    
+    # Override LLM provider from environment if set
+    llm_provider = os.environ.get("LLM_PROVIDER")
+    if llm_provider:
+        config["llm"]["provider"] = llm_provider
+        logging.info(f"LLM provider overridden from environment: {llm_provider}")
+    
+    # Also support direct API key from environment if the provider-specific key isn't set
+    llm_api_key = os.environ.get("LLM_API_KEY")
+    if llm_api_key and "api_key" not in config["llm"]:
+        config["llm"]["api_key"] = llm_api_key
+        logging.info("LLM API key loaded from LLM_API_KEY environment variable")
+    
+    return config
 
 def setup_logging(config: Dict) -> None:
     """Configure application logging."""
@@ -56,6 +71,52 @@ def decode_redis_hash(hash_data: Dict[bytes, bytes]) -> Dict[str, str]:
 
 
 
+def get_llm_api_key(config: Dict, provider: Optional[str] = None) -> str:
+    """Get the appropriate API key based on the LLM provider.
+    
+    Args:
+        config: The application configuration dictionary
+        provider: Optional provider override (e.g., "openai/gpt-4")
+    
+    Returns:
+        The API key for the provider, or empty string if not found
+    """
+        
+    # Use provided provider or fall back to config
+    if not provider:
+        provider = config["llm"]["provider"]
+    
+    # Check if direct API key is configured
+    if "api_key" in config["llm"]:
+        return config["llm"]["api_key"]
+    
+    # Fall back to the configured api_key_env if no match
+    return os.environ.get(config["llm"].get("api_key_env", ""), "")
+
+
+def validate_llm_provider(config: Dict, provider: Optional[str] = None) -> tuple[bool, str]:
+    """Validate that the LLM provider has an associated API key.
+    
+    Args:
+        config: The application configuration dictionary
+        provider: Optional provider override (e.g., "openai/gpt-4")
+    
+    Returns:
+        Tuple of (is_valid, error_message)
+    """
+    # Use provided provider or fall back to config
+    if not provider:
+        provider = config["llm"]["provider"]
+    
+    # Get the API key for this provider
+    api_key = get_llm_api_key(config, provider)
+    
+    if not api_key:
+        return False, f"No API key found for provider '{provider}'. Please set the appropriate environment variable."
+    
+    return True, ""
+
+
 def verify_email_domain(email: str) -> bool:
     try:
         domain = email.split('@')[1]

docker-compose.yml

@@ -14,6 +14,7 @@ x-base-config: &base-config
     - TOGETHER_API_KEY=${TOGETHER_API_KEY:-}
     - MISTRAL_API_KEY=${MISTRAL_API_KEY:-}
     - GEMINI_API_TOKEN=${GEMINI_API_TOKEN:-}
+    - LLM_PROVIDER=${LLM_PROVIDER:-}  # Optional: Override default provider (e.g., "anthropic/claude-3-opus")
   volumes:
     - /dev/shm:/dev/shm  # Chromium performance
   deploy:

docs/blog/release-v0.7.3.md

@@ -0,0 +1,350 @@
+# 🚀 Crawl4AI v0.7.3: The Multi-Config Intelligence Update
+
+*August 6, 2025 • 5 min read*
+
+---
+
+Today I'm releasing Crawl4AI v0.7.3—the Multi-Config Intelligence Update. This release brings smarter URL-specific configurations, flexible Docker deployments, important bug fixes, and documentation improvements that make Crawl4AI more robust and production-ready.
+
+## 🎯 What's New at a Glance
+
+- **🕵️ Undetected Browser Support**: Stealth mode for bypassing bot detection systems
+- **🎨 Multi-URL Configurations**: Different crawling strategies for different URL patterns in a single batch
+- **🐳 Flexible Docker LLM Providers**: Configure LLM providers via environment variables
+- **🧠 Memory Monitoring**: Enhanced memory usage tracking and optimization tools
+- **📊 Enhanced Table Extraction**: Improved table access and DataFrame conversion
+- **💰 GitHub Sponsors**: 4-tier sponsorship system with custom arrangements
+- **🔧 Bug Fixes**: Resolved several critical issues for better stability
+- **📚 Documentation Updates**: Clearer examples and improved API documentation
+
+## 🎨 Multi-URL Configurations: One Size Doesn't Fit All
+
+**The Problem:** You're crawling a mix of documentation sites, blogs, and API endpoints. Each needs different handling—caching for docs, fresh content for news, structured extraction for APIs. Previously, you'd run separate crawls or write complex conditional logic.
+
+**My Solution:** I implemented URL-specific configurations that let you define different strategies for different URL patterns in a single crawl batch. First match wins, with optional fallback support.
+
+### Technical Implementation
+
+```python
+from crawl4ai import AsyncWebCrawler, CrawlerRunConfig, MatchMode
+
+# Define specialized configs for different content types
+configs = [
+    # Documentation sites - aggressive caching, include links
+    CrawlerRunConfig(
+        url_matcher=["*docs*", "*documentation*"],
+        cache_mode="write",
+        markdown_generator_options={"include_links": True}
+    ),
+    
+    # News/blog sites - fresh content, scroll for lazy loading
+    CrawlerRunConfig(
+        url_matcher=lambda url: 'blog' in url or 'news' in url,
+        cache_mode="bypass",
+        js_code="window.scrollTo(0, document.body.scrollHeight/2);"
+    ),
+    
+    # API endpoints - structured extraction
+    CrawlerRunConfig(
+        url_matcher=["*.json", "*api*"],
+        extraction_strategy=LLMExtractionStrategy(
+            provider="openai/gpt-4o-mini",
+            extraction_type="structured"
+        )
+    ),
+    
+    # Default fallback for everything else
+    CrawlerRunConfig()  # No url_matcher = matches everything
+]
+
+# Crawl multiple URLs with appropriate configs
+async with AsyncWebCrawler() as crawler:
+    results = await crawler.arun_many(
+        urls=[
+            "https://docs.python.org/3/",      # → Uses documentation config
+            "https://blog.python.org/",        # → Uses blog config  
+            "https://api.github.com/users",    # → Uses API config
+            "https://example.com/"             # → Uses default config
+        ],
+        config=configs
+    )
+```
+
+**Matching Capabilities:**
+- **String Patterns**: Wildcards like `"*.pdf"`, `"*/blog/*"`
+- **Function Matchers**: Lambda functions for complex logic
+- **Mixed Matchers**: Combine strings and functions with AND/OR logic
+- **Fallback Support**: Default config when nothing matches
+
+**Expected Real-World Impact:**
+- **Mixed Content Sites**: Handle blogs, docs, and downloads in one crawl
+- **Multi-Domain Crawling**: Different strategies per domain without separate runs
+- **Reduced Complexity**: No more if/else forests in your extraction code
+- **Better Performance**: Each URL gets exactly the processing it needs
+
+## 🕵️ Undetected Browser Support: Stealth Mode Activated
+
+**The Problem:** Modern websites employ sophisticated bot detection systems. Cloudflare, Akamai, and custom solutions block automated crawlers, limiting access to valuable content.
+
+**My Solution:** I implemented undetected browser support with a flexible adapter pattern. Now Crawl4AI can bypass most bot detection systems using stealth techniques.
+
+### Technical Implementation
+
+```python
+from crawl4ai import AsyncWebCrawler, BrowserConfig
+
+# Enable undetected mode for stealth crawling
+browser_config = BrowserConfig(
+    browser_type="undetected",  # Use undetected Chrome
+    headless=True,              # Can run headless with stealth
+    extra_args=[
+        "--disable-blink-features=AutomationControlled",
+        "--disable-web-security",
+        "--disable-features=VizDisplayCompositor"
+    ]
+)
+
+async with AsyncWebCrawler(config=browser_config) as crawler:
+    # This will bypass most bot detection systems
+    result = await crawler.arun("https://protected-site.com")
+    
+    if result.success:
+        print("✅ Successfully bypassed bot detection!")
+        print(f"Content length: {len(result.markdown)}")
+```
+
+**Advanced Anti-Bot Strategies:**
+
+```python
+# Combine multiple stealth techniques
+from crawl4ai import CrawlerRunConfig
+
+config = CrawlerRunConfig(
+    # Random user agents and headers
+    headers={
+        "Accept-Language": "en-US,en;q=0.9",
+        "Accept-Encoding": "gzip, deflate, br",
+        "DNT": "1"
+    },
+    
+    # Human-like behavior simulation
+    js_code="""
+        // Random mouse movements
+        const simulateHuman = () => {
+            const event = new MouseEvent('mousemove', {
+                clientX: Math.random() * window.innerWidth,
+                clientY: Math.random() * window.innerHeight
+            });
+            document.dispatchEvent(event);
+        };
+        setInterval(simulateHuman, 100 + Math.random() * 200);
+        
+        // Random scrolling
+        const randomScroll = () => {
+            const scrollY = Math.random() * (document.body.scrollHeight - window.innerHeight);
+            window.scrollTo(0, scrollY);
+        };
+        setTimeout(randomScroll, 500 + Math.random() * 1000);
+    """,
+    
+    # Delay to appear more human
+    delay_before_return_html=2.0
+)
+
+result = await crawler.arun("https://bot-protected-site.com", config=config)
+```
+
+**Expected Real-World Impact:**
+- **Enterprise Scraping**: Access previously blocked corporate sites and databases
+- **Market Research**: Gather data from competitor sites with protection
+- **Price Monitoring**: Track e-commerce sites that block automated access
+- **Content Aggregation**: Collect news and social media despite anti-bot measures
+- **Compliance Testing**: Verify your own site's bot protection effectiveness
+
+## 🧠 Memory Monitoring & Optimization
+
+**The Problem:** Long-running crawl sessions consuming excessive memory, especially when processing large batches or heavy JavaScript sites.
+
+**My Solution:** Built comprehensive memory monitoring and optimization utilities that track usage patterns and provide actionable insights.
+
+### Memory Tracking Implementation
+
+```python
+from crawl4ai.memory_utils import MemoryMonitor, get_memory_info
+
+# Monitor memory during crawling
+monitor = MemoryMonitor()
+
+async with AsyncWebCrawler() as crawler:
+    # Start monitoring
+    monitor.start_monitoring()
+    
+    # Perform memory-intensive operations
+    results = await crawler.arun_many([
+        "https://heavy-js-site.com",
+        "https://large-images-site.com", 
+        "https://dynamic-content-site.com"
+    ])
+    
+    # Get detailed memory report
+    memory_report = monitor.get_report()
+    print(f"Peak memory usage: {memory_report['peak_mb']:.1f} MB")
+    print(f"Memory efficiency: {memory_report['efficiency']:.1f}%")
+    
+    # Automatic cleanup suggestions
+    if memory_report['peak_mb'] > 1000:  # > 1GB
+        print("💡 Consider batch size optimization")
+        print("💡 Enable aggressive garbage collection")
+```
+
+**Expected Real-World Impact:**
+- **Production Stability**: Prevent memory-related crashes in long-running services
+- **Cost Optimization**: Right-size server resources based on actual usage
+- **Performance Tuning**: Identify memory bottlenecks and optimization opportunities
+- **Scalability Planning**: Understand memory patterns for horizontal scaling
+
+## 📊 Enhanced Table Extraction
+
+**The Problem:** Table data was accessed through the generic `result.media` interface, making DataFrame conversion cumbersome and unclear.
+
+**My Solution:** Dedicated `result.tables` interface with direct DataFrame conversion and improved detection algorithms.
+
+### New Table Access Pattern
+
+```python
+# Old way (deprecated)
+# tables_data = result.media.get('tables', [])
+
+# New way (v0.7.3+)
+result = await crawler.arun("https://site-with-tables.com")
+
+# Direct table access
+if result.tables:
+    print(f"Found {len(result.tables)} tables")
+    
+    # Convert to pandas DataFrame instantly
+    import pandas as pd
+    
+    for i, table in enumerate(result.tables):
+        df = pd.DataFrame(table['data'])
+        print(f"Table {i}: {df.shape[0]} rows × {df.shape[1]} columns")
+        print(df.head())
+        
+        # Table metadata
+        print(f"Source: {table.get('source_xpath', 'Unknown')}")
+        print(f"Headers: {table.get('headers', [])}")
+```
+
+**Expected Real-World Impact:**
+- **Data Analysis**: Faster transition from web data to analysis-ready DataFrames
+- **ETL Pipelines**: Cleaner integration with data processing workflows
+- **Reporting**: Simplified table extraction for automated reporting systems
+
+## 💰 Community Support: GitHub Sponsors
+
+I've launched GitHub Sponsors to ensure Crawl4AI's continued development and support our growing community.
+
+**Sponsorship Tiers:**
+- **🌱 Supporter ($5/month)**: Community support + early feature previews
+- **🚀 Professional ($25/month)**: Priority support + beta access
+- **🏢 Business ($100/month)**: Direct consultation + custom integrations
+- **🏛️ Enterprise ($500/month)**: Dedicated support + feature development
+
+**Why Sponsor?**
+- Ensure continuous development and maintenance
+- Get priority support and feature requests
+- Access to premium documentation and examples
+- Direct line to the development team
+
+[**Become a Sponsor →**](https://github.com/sponsors/unclecode)
+
+## 🐳 Docker: Flexible LLM Provider Configuration
+
+**The Problem:** Hardcoded LLM providers in Docker deployments. Want to switch from OpenAI to Groq? Rebuild and redeploy. Testing different models? Multiple Docker images.
+
+**My Solution:** Configure LLM providers via environment variables. Switch providers without touching code or rebuilding images.
+
+### Deployment Flexibility
+
+```bash
+# Option 1: Direct environment variables
+docker run -d \
+  -e LLM_PROVIDER="groq/llama-3.2-3b-preview" \
+  -e GROQ_API_KEY="your-key" \
+  -p 11235:11235 \
+  unclecode/crawl4ai:latest
+
+# Option 2: Using .llm.env file (recommended for production)
+# Create .llm.env file:
+# LLM_PROVIDER=openai/gpt-4o-mini
+# OPENAI_API_KEY=your-openai-key
+# GROQ_API_KEY=your-groq-key
+
+docker run -d \
+  --env-file .llm.env \
+  -p 11235:11235 \
+  unclecode/crawl4ai:latest
+```
+
+Override per request when needed:
+```python
+# Use default provider from .llm.env
+response = requests.post("http://localhost:11235/crawl", json={
+    "url": "https://example.com",
+    "extraction_strategy": {"type": "llm"}
+})
+
+# Override to use different provider for this specific request
+response = requests.post("http://localhost:11235/crawl", json={
+    "url": "https://complex-page.com",
+    "extraction_strategy": {
+        "type": "llm",
+        "provider": "openai/gpt-4"  # Override default
+    }
+})
+```
+
+**Expected Real-World Impact:**
+- **Cost Optimization**: Use cheaper models for simple tasks, premium for complex
+- **A/B Testing**: Compare provider performance without deployment changes
+- **Fallback Strategies**: Switch providers on-the-fly during outages
+- **Development Flexibility**: Test locally with one provider, deploy with another
+- **Secure Configuration**: Keep API keys in `.llm.env` file, not in commands
+
+## 🔧 Bug Fixes & Improvements
+
+This release includes several important bug fixes that improve stability and reliability:
+
+- **URL Matcher Fallback**: Fixed edge cases in URL pattern matching logic
+- **Memory Management**: Resolved memory leaks in long-running crawl sessions
+- **Sitemap Processing**: Fixed redirect handling in sitemap fetching
+- **Table Extraction**: Improved table detection and extraction accuracy
+- **Error Handling**: Better error messages and recovery from network failures
+
+## 📚 Documentation Enhancements
+
+Based on community feedback, we've updated:
+- Clearer examples for multi-URL configuration
+- Improved CrawlResult documentation with all available fields
+- Fixed typos and inconsistencies across documentation
+- Added real-world URLs in examples for better understanding
+- New comprehensive demo showcasing all v0.7.3 features
+
+## 🙏 Acknowledgments
+
+Thanks to our contributors and the entire community for feedback and bug reports.
+
+## 📚 Resources
+
+- [Full Documentation](https://docs.crawl4ai.com)
+- [GitHub Repository](https://github.com/unclecode/crawl4ai)
+- [Discord Community](https://discord.gg/crawl4ai)
+- [Feature Demo](https://github.com/unclecode/crawl4ai/blob/main/docs/releases_review/demo_v0.7.3.py)
+
+---
+
+*Crawl4AI continues to evolve with your needs. This release makes it smarter, more flexible, and more stable. Try the new multi-config feature and flexible Docker deployment—they're game changers!*
+
+**Happy Crawling! 🕷️**
+
+*- The Crawl4AI Team*

docs/examples/c4a_script/api_usage_examples.py

@@ -3,8 +3,8 @@ C4A-Script API Usage Examples
 Shows how to use the new Result-based API in various scenarios
 """
 
-from c4a_compile import compile, validate, compile_file
-from c4a_result import CompilationResult, ValidationResult
+from crawl4ai.script.c4a_compile import compile, validate, compile_file
+from crawl4ai.script.c4a_result import CompilationResult, ValidationResult
 import json
 
 

docs/examples/c4a_script/c4a_script_hello_world.py

@@ -3,7 +3,7 @@ C4A-Script Hello World
 A concise example showing how to use the C4A-Script compiler
 """
 
-from c4a_compile import compile
+from crawl4ai.script.c4a_compile import compile
 
 # Define your C4A-Script
 script = """

docs/examples/c4a_script/c4a_script_hello_world_error.py

@@ -3,7 +3,7 @@ C4A-Script Hello World - Error Example
 Shows how error handling works
 """
 
-from c4a_compile import compile
+from crawl4ai.script.c4a_compile import compile
 
 # Define a script with an error (missing THEN)
 script = """

docs/examples/demo_multi_config_clean.py

@@ -0,0 +1,303 @@
+"""
+🎯 Multi-Config URL Matching Demo
+=================================
+Learn how to use different crawler configurations for different URL patterns
+in a single crawl batch with Crawl4AI's multi-config feature.
+
+Part 1: Understanding URL Matching (Pattern Testing)
+Part 2: Practical Example with Real Crawling
+"""
+
+import asyncio
+from crawl4ai import (
+    AsyncWebCrawler, 
+    CrawlerRunConfig,
+    MatchMode
+)
+from crawl4ai.processors.pdf import PDFContentScrapingStrategy
+from crawl4ai.extraction_strategy import JsonCssExtractionStrategy
+from crawl4ai.content_filter_strategy import PruningContentFilter
+from crawl4ai.markdown_generation_strategy import DefaultMarkdownGenerator
+
+
+def print_section(title):
+    """Print a formatted section header"""
+    print(f"\n{'=' * 60}")
+    print(f"{title}")
+    print(f"{'=' * 60}\n")
+
+
+def test_url_matching(config, test_urls, config_name):
+    """Test URL matching for a config and show results"""
+    print(f"Config: {config_name}")
+    print(f"Matcher: {config.url_matcher}")
+    if hasattr(config, 'match_mode'):
+        print(f"Mode: {config.match_mode.value}")
+    print("-" * 40)
+    
+    for url in test_urls:
+        matches = config.is_match(url)
+        symbol = "✓" if matches else "✗"
+        print(f"{symbol} {url}")
+    print()
+
+
+# ==============================================================================
+# PART 1: Understanding URL Matching
+# ==============================================================================
+
+def demo_part1_pattern_matching():
+    """Part 1: Learn how URL matching works without crawling"""
+    
+    print_section("PART 1: Understanding URL Matching")
+    print("Let's explore different ways to match URLs with configs.\n")
+    
+    # Test URLs we'll use throughout
+    test_urls = [
+        "https://example.com/report.pdf",
+        "https://example.com/data.json",
+        "https://example.com/blog/post-1",
+        "https://example.com/article/news",
+        "https://api.example.com/v1/users",
+        "https://example.com/about"
+    ]
+    
+    # 1.1 Simple String Pattern
+    print("1.1 Simple String Pattern Matching")
+    print("-" * 40)
+    
+    pdf_config = CrawlerRunConfig(
+        url_matcher="*.pdf"
+    )
+    
+    test_url_matching(pdf_config, test_urls, "PDF Config")
+    
+    
+    # 1.2 Multiple String Patterns
+    print("1.2 Multiple String Patterns (OR logic)")
+    print("-" * 40)
+    
+    blog_config = CrawlerRunConfig(
+        url_matcher=["*/blog/*", "*/article/*", "*/news/*"],
+        match_mode=MatchMode.OR  # This is default, shown for clarity
+    )
+    
+    test_url_matching(blog_config, test_urls, "Blog/Article Config")
+    
+    
+    # 1.3 Single Function Matcher
+    print("1.3 Function-based Matching")
+    print("-" * 40)
+    
+    api_config = CrawlerRunConfig(
+        url_matcher=lambda url: 'api' in url or url.endswith('.json')
+    )
+    
+    test_url_matching(api_config, test_urls, "API Config")
+    
+    
+    # 1.4 List of Functions
+    print("1.4 Multiple Functions with AND Logic")
+    print("-" * 40)
+    
+    # Must be HTTPS AND contain 'api' AND have version number
+    secure_api_config = CrawlerRunConfig(
+        url_matcher=[
+            lambda url: url.startswith('https://'),
+            lambda url: 'api' in url,
+            lambda url: '/v' in url  # Version indicator
+        ],
+        match_mode=MatchMode.AND
+    )
+    
+    test_url_matching(secure_api_config, test_urls, "Secure API Config")
+    
+    
+    # 1.5 Mixed: String and Function Together
+    print("1.5 Mixed Patterns: String + Function")
+    print("-" * 40)
+    
+    # Match JSON files OR any API endpoint
+    json_or_api_config = CrawlerRunConfig(
+        url_matcher=[
+            "*.json",  # String pattern
+            lambda url: 'api' in url  # Function
+        ],
+        match_mode=MatchMode.OR
+    )
+    
+    test_url_matching(json_or_api_config, test_urls, "JSON or API Config")
+    
+    
+    # 1.6 Complex: Multiple Strings + Multiple Functions
+    print("1.6 Complex Matcher: Mixed Types with AND Logic")
+    print("-" * 40)
+    
+    # Must be: HTTPS AND (.com domain) AND (blog OR article) AND NOT a PDF
+    complex_config = CrawlerRunConfig(
+        url_matcher=[
+            lambda url: url.startswith('https://'),  # Function: HTTPS check
+            "*.com/*",  # String: .com domain
+            lambda url: any(pattern in url for pattern in ['/blog/', '/article/']),  # Function: Blog OR article
+            lambda url: not url.endswith('.pdf')  # Function: Not PDF
+        ],
+        match_mode=MatchMode.AND
+    )
+    
+    test_url_matching(complex_config, test_urls, "Complex Mixed Config")
+    
+    print("\n✅ Key Takeaway: First matching config wins when passed to arun_many()!")
+
+
+# ==============================================================================
+# PART 2: Practical Multi-URL Crawling
+# ==============================================================================
+
+async def demo_part2_practical_crawling():
+    """Part 2: Real-world example with different content types"""
+    
+    print_section("PART 2: Practical Multi-URL Crawling")
+    print("Now let's see multi-config in action with real URLs.\n")
+    
+    # Create specialized configs for different content types
+    configs = [
+        # Config 1: PDF documents - only match files ending with .pdf
+        CrawlerRunConfig(
+            url_matcher="*.pdf",
+            scraping_strategy=PDFContentScrapingStrategy()
+        ),
+        
+        # Config 2: Blog/article pages with content filtering
+        CrawlerRunConfig(
+            url_matcher=["*/blog/*", "*/article/*", "*python.org*"],
+            markdown_generator=DefaultMarkdownGenerator(
+                content_filter=PruningContentFilter(threshold=0.48)
+            )
+        ),
+        
+        # Config 3: Dynamic pages requiring JavaScript
+        CrawlerRunConfig(
+            url_matcher=lambda url: 'github.com' in url,
+            js_code="window.scrollTo(0, 500);"  # Scroll to load content
+        ),
+        
+        # Config 4: Mixed matcher - API endpoints (string OR function)
+        CrawlerRunConfig(
+            url_matcher=[
+                "*.json",  # String pattern for JSON files
+                lambda url: 'api' in url or 'httpbin.org' in url  # Function for API endpoints
+            ],
+            match_mode=MatchMode.OR,
+        ),
+        
+        # Config 5: Complex matcher - Secure documentation sites
+        CrawlerRunConfig(
+            url_matcher=[
+                lambda url: url.startswith('https://'),  # Must be HTTPS
+                "*.org/*",  # String: .org domain
+                lambda url: any(doc in url for doc in ['docs', 'documentation', 'reference']),  # Has docs
+                lambda url: not url.endswith(('.pdf', '.json'))  # Not PDF or JSON
+            ],
+            match_mode=MatchMode.AND,
+            # wait_for="css:.content, css:article"  # Wait for content to load
+        ),
+        
+        # Default config for everything else
+        # CrawlerRunConfig()  # No url_matcher means it matches everything (use it as fallback)
+    ]
+    
+    # URLs to crawl - each will use a different config
+    urls = [
+        "https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf",  # → PDF config
+        "https://blog.python.org/",  # → Blog config with content filter
+        "https://github.com/microsoft/playwright",  # → JS config
+        "https://httpbin.org/json",  # → Mixed matcher config (API)
+        "https://docs.python.org/3/reference/",  # → Complex matcher config
+        "https://www.w3schools.com/",  # → Default config, if you uncomment the default config line above, if not you will see `Error: No matching configuration`
+    ]
+    
+    print("URLs to crawl:")
+    for i, url in enumerate(urls, 1):
+        print(f"{i}. {url}")
+    
+    print("\nCrawling with appropriate config for each URL...\n")
+    
+    async with AsyncWebCrawler() as crawler:
+        results = await crawler.arun_many(
+            urls=urls,
+            config=configs
+        )
+        
+        # Display results
+        print("Results:")
+        print("-" * 60)
+        
+        for result in results:
+            if result.success:
+                # Determine which config was used
+                config_type = "Default"
+                if result.url.endswith('.pdf'):
+                    config_type = "PDF Strategy"
+                elif any(pattern in result.url for pattern in ['blog', 'python.org']) and 'docs' not in result.url:
+                    config_type = "Blog + Content Filter"
+                elif 'github.com' in result.url:
+                    config_type = "JavaScript Enabled"
+                elif 'httpbin.org' in result.url or result.url.endswith('.json'):
+                    config_type = "Mixed Matcher (API)"
+                elif 'docs.python.org' in result.url:
+                    config_type = "Complex Matcher (Secure Docs)"
+                
+                print(f"\n✓ {result.url}")
+                print(f"  Config used: {config_type}")
+                print(f"  Content size: {len(result.markdown)} chars")
+                
+                # Show if we have fit_markdown (from content filter)
+                if hasattr(result.markdown, 'fit_markdown') and result.markdown.fit_markdown:
+                    print(f"  Fit markdown size: {len(result.markdown.fit_markdown)} chars")
+                    reduction = (1 - len(result.markdown.fit_markdown) / len(result.markdown)) * 100
+                    print(f"  Content reduced by: {reduction:.1f}%")
+                
+                # Show extracted data if using extraction strategy
+                if hasattr(result, 'extracted_content') and result.extracted_content:
+                    print(f"  Extracted data: {str(result.extracted_content)[:100]}...")
+            else:
+                print(f"\n✗ {result.url}")
+                print(f"  Error: {result.error_message}")
+    
+    print("\n" + "=" * 60)
+    print("✅ Multi-config crawling complete!")
+    print("\nBenefits demonstrated:")
+    print("- PDFs handled with specialized scraper")
+    print("- Blog content filtered for relevance") 
+    print("- JavaScript executed only where needed")
+    print("- Mixed matchers (string + function) for flexible matching")
+    print("- Complex matchers for precise URL targeting")
+    print("- Each URL got optimal configuration automatically!")
+
+
+async def main():
+    """Run both parts of the demo"""
+    
+    print("""
+🎯 Multi-Config URL Matching Demo
+=================================
+Learn how Crawl4AI can use different configurations
+for different URLs in a single batch.
+    """)
+    
+    # Part 1: Pattern matching
+    demo_part1_pattern_matching()
+    
+    print("\nPress Enter to continue to Part 2...")
+    try:
+        input()
+    except EOFError:
+        # Running in non-interactive mode, skip input
+        pass
+    
+    # Part 2: Practical crawling
+    await demo_part2_practical_crawling()
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

docs/examples/docker_example.py

@@ -8,26 +8,20 @@ from typing import Dict, Any
 
 
 class Crawl4AiTester:
-    def __init__(self, base_url: str = "http://localhost:11235", api_token: str = None):
+    def __init__(self, base_url: str = "http://localhost:11235"):
         self.base_url = base_url
-        self.api_token = (
-            api_token or os.getenv("CRAWL4AI_API_TOKEN") or "test_api_code"
-        )  # Check environment variable as fallback
-        self.headers = (
-            {"Authorization": f"Bearer {self.api_token}"} if self.api_token else {}
-        )
 
     def submit_and_wait(
         self, request_data: Dict[str, Any], timeout: int = 300
     ) -> Dict[str, Any]:
-        # Submit crawl job
+        # Submit crawl job using async endpoint
         response = requests.post(
-            f"{self.base_url}/crawl", json=request_data, headers=self.headers
+            f"{self.base_url}/crawl/job", json=request_data
         )
-        if response.status_code == 403:
-            raise Exception("API token is invalid or missing")
-        task_id = response.json()["task_id"]
-        print(f"Task ID: {task_id}")
+        response.raise_for_status()
+        job_response = response.json()
+        task_id = job_response["task_id"]
+        print(f"Submitted job with task_id: {task_id}")
 
         # Poll for result
         start_time = time.time()
@@ -38,8 +32,9 @@ class Crawl4AiTester:
                 )
 
             result = requests.get(
-                f"{self.base_url}/task/{task_id}", headers=self.headers
+                f"{self.base_url}/crawl/job/{task_id}"
             )
+            result.raise_for_status()
             status = result.json()
 
             if status["status"] == "failed":
@@ -52,10 +47,10 @@ class Crawl4AiTester:
             time.sleep(2)
 
     def submit_sync(self, request_data: Dict[str, Any]) -> Dict[str, Any]:
+        # Use synchronous crawl endpoint
         response = requests.post(
-            f"{self.base_url}/crawl_sync",
+            f"{self.base_url}/crawl",
             json=request_data,
-            headers=self.headers,
             timeout=60,
         )
         if response.status_code == 408:
@@ -63,20 +58,9 @@ class Crawl4AiTester:
         response.raise_for_status()
         return response.json()
 
-    def crawl_direct(self, request_data: Dict[str, Any]) -> Dict[str, Any]:
-        """Directly crawl without using task queue"""
-        response = requests.post(
-            f"{self.base_url}/crawl_direct", json=request_data, headers=self.headers
-        )
-        response.raise_for_status()
-        return response.json()
-
-
 def test_docker_deployment(version="basic"):
     tester = Crawl4AiTester(
         base_url="http://localhost:11235",
-        # base_url="https://api.crawl4ai.com" # just for example
-        # api_token="test" # just for example
     )
     print(f"Testing Crawl4AI Docker {version} version")
 
@@ -95,11 +79,8 @@ def test_docker_deployment(version="basic"):
             time.sleep(5)
 
     # Test cases based on version
-    test_basic_crawl_direct(tester)
-    test_basic_crawl(tester)
     test_basic_crawl(tester)
     test_basic_crawl_sync(tester)
-
     if version in ["full", "transformer"]:
         test_cosine_extraction(tester)
 
@@ -112,115 +93,129 @@ def test_docker_deployment(version="basic"):
 
 
 def test_basic_crawl(tester: Crawl4AiTester):
-    print("\n=== Testing Basic Crawl ===")
+    print("\n=== Testing Basic Crawl (Async) ===")
     request = {
-        "urls": "https://www.nbcnews.com/business",
-        "priority": 10,
-        "session_id": "test",
+        "urls": ["https://www.nbcnews.com/business"],
+        "browser_config": {},
+        "crawler_config": {}
     }
 
     result = tester.submit_and_wait(request)
-    print(f"Basic crawl result length: {len(result['result']['markdown'])}")
+    print(f"Basic crawl result count: {len(result['result']['results'])}")
     assert result["result"]["success"]
-    assert len(result["result"]["markdown"]) > 0
+    assert len(result["result"]["results"]) > 0
+    assert len(result["result"]["results"][0]["markdown"]) > 0
 
 
 def test_basic_crawl_sync(tester: Crawl4AiTester):
     print("\n=== Testing Basic Crawl (Sync) ===")
     request = {
-        "urls": "https://www.nbcnews.com/business",
-        "priority": 10,
-        "session_id": "test",
+        "urls": ["https://www.nbcnews.com/business"],
+        "browser_config": {},
+        "crawler_config": {}
     }
 
     result = tester.submit_sync(request)
-    print(f"Basic crawl result length: {len(result['result']['markdown'])}")
-    assert result["status"] == "completed"
-    assert result["result"]["success"]
-    assert len(result["result"]["markdown"]) > 0
-
-
-def test_basic_crawl_direct(tester: Crawl4AiTester):
-    print("\n=== Testing Basic Crawl (Direct) ===")
-    request = {
-        "urls": "https://www.nbcnews.com/business",
-        "priority": 10,
-        # "session_id": "test"
-        "cache_mode": "bypass",  # or "enabled", "disabled", "read_only", "write_only"
-    }
-
-    result = tester.crawl_direct(request)
-    print(f"Basic crawl result length: {len(result['result']['markdown'])}")
-    assert result["result"]["success"]
-    assert len(result["result"]["markdown"]) > 0
+    print(f"Basic crawl result count: {len(result['results'])}")
+    assert result["success"]
+    assert len(result["results"]) > 0
+    assert len(result["results"][0]["markdown"]) > 0
 
 
 def test_js_execution(tester: Crawl4AiTester):
     print("\n=== Testing JS Execution ===")
     request = {
-        "urls": "https://www.nbcnews.com/business",
-        "priority": 8,
-        "js_code": [
-            "const loadMoreButton = Array.from(document.querySelectorAll('button')).find(button => button.textContent.includes('Load More')); loadMoreButton && loadMoreButton.click();"
-        ],
-        "wait_for": "article.tease-card:nth-child(10)",
-        "crawler_params": {"headless": True},
+        "urls": ["https://www.nbcnews.com/business"],
+        "browser_config": {"headless": True},
+        "crawler_config": {
+            "js_code": [
+                "const loadMoreButton = Array.from(document.querySelectorAll('button')).find(button => button.textContent.includes('Load More')); if(loadMoreButton) loadMoreButton.click();"
+            ],
+            "wait_for": "wide-tease-item__wrapper df flex-column flex-row-m flex-nowrap-m enable-new-sports-feed-mobile-design(10)"
+        }
     }
 
     result = tester.submit_and_wait(request)
-    print(f"JS execution result length: {len(result['result']['markdown'])}")
+    print(f"JS execution result count: {len(result['result']['results'])}")
     assert result["result"]["success"]
 
 
 def test_css_selector(tester: Crawl4AiTester):
     print("\n=== Testing CSS Selector ===")
     request = {
-        "urls": "https://www.nbcnews.com/business",
-        "priority": 7,
-        "css_selector": ".wide-tease-item__description",
-        "crawler_params": {"headless": True},
-        "extra": {"word_count_threshold": 10},
+        "urls": ["https://www.nbcnews.com/business"],
+        "browser_config": {"headless": True},
+        "crawler_config": {
+            "css_selector": ".wide-tease-item__description",
+            "word_count_threshold": 10
+        }
     }
 
     result = tester.submit_and_wait(request)
-    print(f"CSS selector result length: {len(result['result']['markdown'])}")
+    print(f"CSS selector result count: {len(result['result']['results'])}")
     assert result["result"]["success"]
 
 
 def test_structured_extraction(tester: Crawl4AiTester):
     print("\n=== Testing Structured Extraction ===")
     schema = {
-        "name": "Coinbase Crypto Prices",
-        "baseSelector": ".cds-tableRow-t45thuk",
+        "name": "Cryptocurrency Prices",
+        "baseSelector": "table[data-testid=\"prices-table\"] tbody tr",
         "fields": [
             {
-                "name": "crypto",
-                "selector": "td:nth-child(1) h2",
-                "type": "text",
+                "name": "asset_name",
+                "selector": "td:nth-child(2) p.cds-headline-h4steop",
+                "type": "text"
             },
             {
-                "name": "symbol",
-                "selector": "td:nth-child(1) p",
-                "type": "text",
+                "name": "asset_symbol",
+                "selector": "td:nth-child(2) p.cds-label2-l1sm09ec",
+                "type": "text"
+            },
+            {
+                "name": "asset_image_url",
+                "selector": "td:nth-child(2) img[alt=\"Asset Symbol\"]",
+                "type": "attribute",
+                "attribute": "src"
+            },
+            {
+                "name": "asset_url",
+                "selector": "td:nth-child(2) a[aria-label^=\"Asset page for\"]",
+                "type": "attribute",
+                "attribute": "href"
             },
             {
                 "name": "price",
-                "selector": "td:nth-child(2)",
-                "type": "text",
+                "selector": "td:nth-child(3) div.cds-typographyResets-t6muwls.cds-body-bwup3gq",
+                "type": "text"
             },
-        ],
+            {
+                "name": "change",
+                "selector": "td:nth-child(7) p.cds-body-bwup3gq",
+                "type": "text"
+            }
+        ]
     }
 
     request = {
-        "urls": "https://www.coinbase.com/explore",
-        "priority": 9,
-        "extraction_config": {"type": "json_css", "params": {"schema": schema}},
+        "urls": ["https://www.coinbase.com/explore"],
+        "browser_config": {},
+        "crawler_config": {
+            "type": "CrawlerRunConfig",
+            "params": {
+                "extraction_strategy": {
+                    "type": "JsonCssExtractionStrategy",
+                    "params": {"schema": schema}
+                }
+            }
+        }
     }
 
     result = tester.submit_and_wait(request)
-    extracted = json.loads(result["result"]["extracted_content"])
+    extracted = json.loads(result["result"]["results"][0]["extracted_content"])
     print(f"Extracted {len(extracted)} items")
-    print("Sample item:", json.dumps(extracted[0], indent=2))
+    if extracted:
+        print("Sample item:", json.dumps(extracted[0], indent=2))
     assert result["result"]["success"]
     assert len(extracted) > 0
 
@@ -230,43 +225,54 @@ def test_llm_extraction(tester: Crawl4AiTester):
     schema = {
         "type": "object",
         "properties": {
-            "model_name": {
+            "asset_name": {
                 "type": "string",
-                "description": "Name of the OpenAI model.",
+                "description": "Name of the asset.",
             },
-            "input_fee": {
+            "price": {
                 "type": "string",
-                "description": "Fee for input token for the OpenAI model.",
+                "description": "Price of the asset.",
             },
-            "output_fee": {
+            "change": {
                 "type": "string",
-                "description": "Fee for output token for the OpenAI model.",
+                "description": "Change in price of the asset.",
             },
         },
-        "required": ["model_name", "input_fee", "output_fee"],
+        "required": ["asset_name", "price", "change"],
     }
 
     request = {
-        "urls": "https://openai.com/api/pricing",
-        "priority": 8,
-        "extraction_config": {
-            "type": "llm",
+        "urls": ["https://www.coinbase.com/en-in/explore"],
+        "browser_config": {},
+        "crawler_config": {
+            "type": "CrawlerRunConfig",
             "params": {
-                "provider": "openai/gpt-4o-mini",
-                "api_token": os.getenv("OPENAI_API_KEY"),
-                "schema": schema,
-                "extraction_type": "schema",
-                "instruction": """From the crawled content, extract all mentioned model names along with their fees for input and output tokens.""",
-            },
-        },
-        "crawler_params": {"word_count_threshold": 1},
+                "extraction_strategy": {
+                    "type": "LLMExtractionStrategy",
+                    "params": {
+                        "llm_config": {
+                            "type": "LLMConfig",
+                            "params": {
+                                "provider": "gemini/gemini-2.0-flash-exp",
+                                "api_token": os.getenv("GEMINI_API_KEY")
+                            }
+                        },
+                        "schema": schema,
+                        "extraction_type": "schema",
+                        "instruction": "From the crawled content, extract asset names along with their prices and change in price.",
+                    }
+                },
+                "word_count_threshold": 1
+            }
+        }
     }
 
     try:
         result = tester.submit_and_wait(request)
-        extracted = json.loads(result["result"]["extracted_content"])
-        print(f"Extracted {len(extracted)} model pricing entries")
-        print("Sample entry:", json.dumps(extracted[0], indent=2))
+        extracted = json.loads(result["result"]["results"][0]["extracted_content"])
+        print(f"Extracted {len(extracted)} asset pricing entries")
+        if extracted:
+            print("Sample entry:", json.dumps(extracted[0], indent=2))
         assert result["result"]["success"]
     except Exception as e:
         print(f"LLM extraction test failed (might be due to missing API key): {str(e)}")
@@ -274,6 +280,16 @@ def test_llm_extraction(tester: Crawl4AiTester):
 
 def test_llm_with_ollama(tester: Crawl4AiTester):
     print("\n=== Testing LLM with Ollama ===")
+    
+    # Check if Ollama is accessible first
+    try:
+        ollama_response = requests.get("http://localhost:11434/api/tags", timeout=5)
+        ollama_response.raise_for_status()
+        print("Ollama is accessible")
+    except:
+        print("Ollama is not accessible, skipping test")
+        return
+    
     schema = {
         "type": "object",
         "properties": {
@@ -294,24 +310,33 @@ def test_llm_with_ollama(tester: Crawl4AiTester):
     }
 
     request = {
-        "urls": "https://www.nbcnews.com/business",
-        "priority": 8,
-        "extraction_config": {
-            "type": "llm",
+        "urls": ["https://www.nbcnews.com/business"],
+        "browser_config": {"verbose": True},
+        "crawler_config": {
+            "type": "CrawlerRunConfig",
             "params": {
-                "provider": "ollama/llama2",
-                "schema": schema,
-                "extraction_type": "schema",
-                "instruction": "Extract the main article information including title, summary, and main topics.",
-            },
-        },
-        "extra": {"word_count_threshold": 1},
-        "crawler_params": {"verbose": True},
+                "extraction_strategy": {
+                    "type": "LLMExtractionStrategy",
+                    "params": {
+                        "llm_config": {
+                            "type": "LLMConfig",
+                            "params": {
+                                "provider": "ollama/llama3.2:latest",
+                            }
+                        },
+                        "schema": schema,
+                        "extraction_type": "schema",
+                        "instruction": "Extract the main article information including title, summary, and main topics.",
+                    }
+                },
+                "word_count_threshold": 1
+            }
+        }
     }
 
     try:
         result = tester.submit_and_wait(request)
-        extracted = json.loads(result["result"]["extracted_content"])
+        extracted = json.loads(result["result"]["results"][0]["extracted_content"])
         print("Extracted content:", json.dumps(extracted, indent=2))
         assert result["result"]["success"]
     except Exception as e:
@@ -321,24 +346,30 @@ def test_llm_with_ollama(tester: Crawl4AiTester):
 def test_cosine_extraction(tester: Crawl4AiTester):
     print("\n=== Testing Cosine Extraction ===")
     request = {
-        "urls": "https://www.nbcnews.com/business",
-        "priority": 8,
-        "extraction_config": {
-            "type": "cosine",
+        "urls": ["https://www.nbcnews.com/business"],
+        "browser_config": {},
+        "crawler_config": {
+            "type": "CrawlerRunConfig",
             "params": {
-                "semantic_filter": "business finance economy",
-                "word_count_threshold": 10,
-                "max_dist": 0.2,
-                "top_k": 3,
-            },
-        },
+                "extraction_strategy": {
+                    "type": "CosineStrategy",
+                    "params": {
+                        "semantic_filter": "business finance economy",
+                        "word_count_threshold": 10,
+                        "max_dist": 0.2,
+                        "top_k": 3,
+                    }
+                }
+            }
+        }
     }
 
     try:
         result = tester.submit_and_wait(request)
-        extracted = json.loads(result["result"]["extracted_content"])
+        extracted = json.loads(result["result"]["results"][0]["extracted_content"])
         print(f"Extracted {len(extracted)} text clusters")
-        print("First cluster tags:", extracted[0]["tags"])
+        if extracted:
+            print("First cluster tags:", extracted[0]["tags"])
         assert result["result"]["success"]
     except Exception as e:
         print(f"Cosine extraction test failed: {str(e)}")
@@ -347,20 +378,25 @@ def test_cosine_extraction(tester: Crawl4AiTester):
 def test_screenshot(tester: Crawl4AiTester):
     print("\n=== Testing Screenshot ===")
     request = {
-        "urls": "https://www.nbcnews.com/business",
-        "priority": 5,
-        "screenshot": True,
-        "crawler_params": {"headless": True},
+        "urls": ["https://www.nbcnews.com/business"],
+        "browser_config": {"headless": True},
+        "crawler_config": {
+            "type": "CrawlerRunConfig",
+            "params": {
+                "screenshot": True
+            }
+        }
     }
 
     result = tester.submit_and_wait(request)
-    print("Screenshot captured:", bool(result["result"]["screenshot"]))
+    screenshot_data = result["result"]["results"][0]["screenshot"]
+    print("Screenshot captured:", bool(screenshot_data))
 
-    if result["result"]["screenshot"]:
+    if screenshot_data:
         # Save screenshot
-        screenshot_data = base64.b64decode(result["result"]["screenshot"])
+        screenshot_bytes = base64.b64decode(screenshot_data)
         with open("test_screenshot.jpg", "wb") as f:
-            f.write(screenshot_data)
+            f.write(screenshot_bytes)
         print("Screenshot saved as test_screenshot.jpg")
 
     assert result["result"]["success"]
@@ -368,5 +404,4 @@ def test_screenshot(tester: Crawl4AiTester):
 
 if __name__ == "__main__":
     version = sys.argv[1] if len(sys.argv) > 1 else "basic"
-    # version = "full"
     test_docker_deployment(version)

docs/examples/hello_world_undetected.py

@@ -0,0 +1,57 @@
+import asyncio
+from crawl4ai import (
+    AsyncWebCrawler,
+    BrowserConfig,
+    CrawlerRunConfig,
+    DefaultMarkdownGenerator,
+    PruningContentFilter,
+    CrawlResult,
+    UndetectedAdapter
+)
+from crawl4ai.async_crawler_strategy import AsyncPlaywrightCrawlerStrategy
+
+
+async def main():
+    # Create browser config
+    browser_config = BrowserConfig(
+        headless=False,
+        verbose=True,
+    )
+    
+    # Create the undetected adapter
+    undetected_adapter = UndetectedAdapter()
+    
+    # Create the crawler strategy with the undetected adapter
+    crawler_strategy = AsyncPlaywrightCrawlerStrategy(
+        browser_config=browser_config,
+        browser_adapter=undetected_adapter
+    )
+    
+    # Create the crawler with our custom strategy
+    async with AsyncWebCrawler(
+        crawler_strategy=crawler_strategy,
+        config=browser_config
+    ) as crawler:
+        # Configure the crawl
+        crawler_config = CrawlerRunConfig(
+            markdown_generator=DefaultMarkdownGenerator(
+                content_filter=PruningContentFilter()
+            ),
+            capture_console_messages=True,  # Enable console capture to test adapter
+        )
+        
+        # Test on a site that typically detects bots
+        print("Testing undetected adapter...")
+        result: CrawlResult = await crawler.arun(
+            url="https://www.helloworld.org", 
+            config=crawler_config
+        )
+        
+        print(f"Status: {result.status_code}")
+        print(f"Success: {result.success}")
+        print(f"Console messages captured: {len(result.console_messages or [])}")
+        print(f"Markdown content (first 500 chars):\n{result.markdown.raw_markdown[:500]}")
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

docs/examples/scraping_strategies_performance.py

@@ -1,5 +1,6 @@
 import time, re
-from crawl4ai.content_scraping_strategy import WebScrapingStrategy,  LXMLWebScrapingStrategy
+from crawl4ai.content_scraping_strategy import LXMLWebScrapingStrategy
+# WebScrapingStrategy is now an alias for LXMLWebScrapingStrategy
 import time
 import functools
 from collections import defaultdict
@@ -57,7 +58,7 @@ methods_to_profile = [
 
 
 # Apply decorators to both strategies
-for strategy, name in [(WebScrapingStrategy, "Original"), (LXMLWebScrapingStrategy, "LXML")]:
+for strategy, name in [(LXMLWebScrapingStrategy, "LXML")]:
     for method in methods_to_profile:
         apply_decorators(strategy, method, name)
 
@@ -85,7 +86,7 @@ def generate_large_html(n_elements=1000):
 
 def test_scraping():
     # Initialize both scrapers
-    original_scraper = WebScrapingStrategy()
+    original_scraper = LXMLWebScrapingStrategy()
     selected_scraper = LXMLWebScrapingStrategy()
     
     # Generate test HTML

docs/examples/simple_anti_bot_examples.py

@@ -0,0 +1,59 @@
+import asyncio
+from crawl4ai import AsyncWebCrawler, BrowserConfig, CrawlerRunConfig, UndetectedAdapter
+from crawl4ai.async_crawler_strategy import AsyncPlaywrightCrawlerStrategy
+
+# Example 1: Stealth Mode
+async def stealth_mode_example():
+    browser_config = BrowserConfig(
+        enable_stealth=True,
+        headless=False
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        result = await crawler.arun("https://example.com")
+        return result.html[:500]
+
+# Example 2: Undetected Browser
+async def undetected_browser_example():
+    browser_config = BrowserConfig(
+        headless=False
+    )
+    
+    adapter = UndetectedAdapter()
+    strategy = AsyncPlaywrightCrawlerStrategy(
+        browser_config=browser_config,
+        browser_adapter=adapter
+    )
+    
+    async with AsyncWebCrawler(
+        crawler_strategy=strategy,
+        config=browser_config
+    ) as crawler:
+        result = await crawler.arun("https://example.com")
+        return result.html[:500]
+
+# Example 3: Both Combined
+async def combined_example():
+    browser_config = BrowserConfig(
+        enable_stealth=True,
+        headless=False
+    )
+    
+    adapter = UndetectedAdapter()
+    strategy = AsyncPlaywrightCrawlerStrategy(
+        browser_config=browser_config,
+        browser_adapter=adapter
+    )
+    
+    async with AsyncWebCrawler(
+        crawler_strategy=strategy,
+        config=browser_config
+    ) as crawler:
+        result = await crawler.arun("https://example.com")
+        return result.html[:500]
+
+# Run examples
+if __name__ == "__main__":
+    asyncio.run(stealth_mode_example())
+    asyncio.run(undetected_browser_example())
+    asyncio.run(combined_example())

docs/examples/stealth_mode_example.py

@@ -0,0 +1,522 @@
+"""
+Stealth Mode Example with Crawl4AI
+
+This example demonstrates how to use the stealth mode feature to bypass basic bot detection.
+The stealth mode uses playwright-stealth to modify browser fingerprints and behaviors
+that are commonly used to detect automated browsers.
+
+Key features demonstrated:
+1. Comparing crawling with and without stealth mode
+2. Testing against bot detection sites
+3. Accessing sites that block automated browsers
+4. Best practices for stealth crawling
+"""
+
+import asyncio
+import json
+from typing import Dict, Any
+from colorama import Fore, Style, init
+
+from crawl4ai import AsyncWebCrawler, BrowserConfig, CrawlerRunConfig
+from crawl4ai.async_logger import AsyncLogger
+
+# Initialize colorama for colored output
+init()
+
+# Create a logger for better output
+logger = AsyncLogger(verbose=True)
+
+
+async def test_bot_detection(use_stealth: bool = False) -> Dict[str, Any]:
+    """Test against a bot detection service"""
+    
+    logger.info(
+        f"Testing bot detection with stealth={'ON' if use_stealth else 'OFF'}",
+        tag="STEALTH"
+    )
+    
+    # Configure browser with or without stealth
+    browser_config = BrowserConfig(
+        headless=False,  # Use False to see the browser in action
+        enable_stealth=use_stealth,
+        viewport_width=1280,
+        viewport_height=800
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        # JavaScript to extract bot detection results
+        detection_script = """
+        // Comprehensive bot detection checks
+        (() => {
+        const detectionResults = {
+            // Basic WebDriver detection
+            webdriver: navigator.webdriver,
+            
+            // Chrome specific
+            chrome: !!window.chrome,
+            chromeRuntime: !!window.chrome?.runtime,
+            
+            // Automation indicators
+            automationControlled: navigator.webdriver,
+            
+            // Permissions API
+            permissionsPresent: !!navigator.permissions?.query,
+            
+            // Plugins
+            pluginsLength: navigator.plugins.length,
+            pluginsArray: Array.from(navigator.plugins).map(p => p.name),
+            
+            // Languages
+            languages: navigator.languages,
+            language: navigator.language,
+            
+            // User agent
+            userAgent: navigator.userAgent,
+            
+            // Screen and window properties
+            screen: {
+                width: screen.width,
+                height: screen.height,
+                availWidth: screen.availWidth,
+                availHeight: screen.availHeight,
+                colorDepth: screen.colorDepth,
+                pixelDepth: screen.pixelDepth
+            },
+            
+            // WebGL vendor
+            webglVendor: (() => {
+                try {
+                    const canvas = document.createElement('canvas');
+                    const gl = canvas.getContext('webgl') || canvas.getContext('experimental-webgl');
+                    const ext = gl.getExtension('WEBGL_debug_renderer_info');
+                    return gl.getParameter(ext.UNMASKED_VENDOR_WEBGL);
+                } catch (e) {
+                    return 'Error';
+                }
+            })(),
+            
+            // Platform
+            platform: navigator.platform,
+            
+            // Hardware concurrency
+            hardwareConcurrency: navigator.hardwareConcurrency,
+            
+            // Device memory
+            deviceMemory: navigator.deviceMemory,
+            
+            // Connection
+            connection: navigator.connection?.effectiveType
+        };
+        
+        // Log results for console capture
+        console.log('DETECTION_RESULTS:', JSON.stringify(detectionResults, null, 2));
+        
+        // Return results
+        return detectionResults;
+        })();
+        """
+        
+        # Crawl bot detection test page
+        config = CrawlerRunConfig(
+            js_code=detection_script,
+            capture_console_messages=True,
+            wait_until="networkidle",
+            delay_before_return_html=2.0  # Give time for all checks to complete
+        )
+        
+        result = await crawler.arun(
+            url="https://bot.sannysoft.com",
+            config=config
+        )
+        
+        if result.success:
+            # Extract detection results from console
+            detection_data = None
+            for msg in result.console_messages or []:
+                if "DETECTION_RESULTS:" in msg.get("text", ""):
+                    try:
+                        json_str = msg["text"].replace("DETECTION_RESULTS:", "").strip()
+                        detection_data = json.loads(json_str)
+                    except:
+                        pass
+            
+            # Also try to get from JavaScript execution result
+            if not detection_data and result.js_execution_result:
+                detection_data = result.js_execution_result
+            
+            return {
+                "success": True,
+                "url": result.url,
+                "detection_data": detection_data,
+                "page_title": result.metadata.get("title", ""),
+                "stealth_enabled": use_stealth
+            }
+        else:
+            return {
+                "success": False,
+                "error": result.error_message,
+                "stealth_enabled": use_stealth
+            }
+
+
+async def test_cloudflare_site(use_stealth: bool = False) -> Dict[str, Any]:
+    """Test accessing a Cloudflare-protected site"""
+    
+    logger.info(
+        f"Testing Cloudflare site with stealth={'ON' if use_stealth else 'OFF'}",
+        tag="STEALTH"
+    )
+    
+    browser_config = BrowserConfig(
+        headless=True,  # Cloudflare detection works better in headless mode with stealth
+        enable_stealth=use_stealth,
+        viewport_width=1920,
+        viewport_height=1080
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        config = CrawlerRunConfig(
+            wait_until="networkidle",
+            page_timeout=30000,  # 30 seconds
+            delay_before_return_html=3.0
+        )
+        
+        # Test on a site that often shows Cloudflare challenges
+        result = await crawler.arun(
+            url="https://nowsecure.nl",
+            config=config
+        )
+        
+        # Check if we hit Cloudflare challenge
+        cloudflare_detected = False
+        if result.html:
+            cloudflare_indicators = [
+                "Checking your browser",
+                "Just a moment",
+                "cf-browser-verification",
+                "cf-challenge",
+                "ray ID"
+            ]
+            cloudflare_detected = any(indicator in result.html for indicator in cloudflare_indicators)
+        
+        return {
+            "success": result.success,
+            "url": result.url,
+            "cloudflare_challenge": cloudflare_detected,
+            "status_code": result.status_code,
+            "page_title": result.metadata.get("title", "") if result.metadata else "",
+            "stealth_enabled": use_stealth,
+            "html_snippet": result.html[:500] if result.html else ""
+        }
+
+
+async def test_anti_bot_site(use_stealth: bool = False) -> Dict[str, Any]:
+    """Test against sites with anti-bot measures"""
+    
+    logger.info(
+        f"Testing anti-bot site with stealth={'ON' if use_stealth else 'OFF'}",
+        tag="STEALTH"
+    )
+    
+    browser_config = BrowserConfig(
+        headless=False,
+        enable_stealth=use_stealth,
+        # Additional browser arguments that help with stealth
+        extra_args=[
+            "--disable-blink-features=AutomationControlled",
+            "--disable-features=site-per-process"
+        ] if not use_stealth else []  # These are automatically applied with stealth
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        # Some sites check for specific behaviors
+        behavior_script = """
+        (async () => {
+            // Simulate human-like behavior
+            const sleep = ms => new Promise(resolve => setTimeout(resolve, ms));
+            
+            // Random mouse movement
+            const moveX = Math.random() * 100;
+            const moveY = Math.random() * 100;
+            
+            // Simulate reading time
+            await sleep(1000 + Math.random() * 2000);
+            
+            // Scroll slightly
+            window.scrollBy(0, 100 + Math.random() * 200);
+            
+            console.log('Human behavior simulation complete');
+            return true;
+        })()
+        """
+        
+        config = CrawlerRunConfig(
+            js_code=behavior_script,
+            wait_until="networkidle",
+            delay_before_return_html=5.0,  # Longer delay to appear more human
+            capture_console_messages=True
+        )
+        
+        # Test on a site that implements anti-bot measures
+        result = await crawler.arun(
+            url="https://www.g2.com/",
+            config=config
+        )
+        
+        # Check for common anti-bot blocks
+        blocked_indicators = [
+            "Access Denied",
+            "403 Forbidden", 
+            "Security Check",
+            "Verify you are human",
+            "captcha",
+            "challenge"
+        ]
+        
+        blocked = False
+        if result.html:
+            blocked = any(indicator.lower() in result.html.lower() for indicator in blocked_indicators)
+        
+        return {
+            "success": result.success and not blocked,
+            "url": result.url,
+            "blocked": blocked,
+            "status_code": result.status_code,
+            "page_title": result.metadata.get("title", "") if result.metadata else "",
+            "stealth_enabled": use_stealth
+        }
+
+
+async def compare_results():
+    """Run all tests with and without stealth mode and compare results"""
+    
+    print(f"\n{Fore.CYAN}{'='*60}{Style.RESET_ALL}")
+    print(f"{Fore.CYAN}Crawl4AI Stealth Mode Comparison{Style.RESET_ALL}")
+    print(f"{Fore.CYAN}{'='*60}{Style.RESET_ALL}\n")
+    
+    # Test 1: Bot Detection
+    print(f"{Fore.YELLOW}1. Bot Detection Test (bot.sannysoft.com){Style.RESET_ALL}")
+    print("-" * 40)
+    
+    # Without stealth
+    regular_detection = await test_bot_detection(use_stealth=False)
+    if regular_detection["success"] and regular_detection["detection_data"]:
+        print(f"{Fore.RED}Without Stealth:{Style.RESET_ALL}")
+        data = regular_detection["detection_data"]
+        print(f"  • WebDriver detected: {data.get('webdriver', 'Unknown')}")
+        print(f"  • Chrome: {data.get('chrome', 'Unknown')}")
+        print(f"  • Languages: {data.get('languages', 'Unknown')}")
+        print(f"  • Plugins: {data.get('pluginsLength', 'Unknown')}")
+        print(f"  • User Agent: {data.get('userAgent', 'Unknown')[:60]}...")
+    
+    # With stealth
+    stealth_detection = await test_bot_detection(use_stealth=True)
+    if stealth_detection["success"] and stealth_detection["detection_data"]:
+        print(f"\n{Fore.GREEN}With Stealth:{Style.RESET_ALL}")
+        data = stealth_detection["detection_data"]
+        print(f"  • WebDriver detected: {data.get('webdriver', 'Unknown')}")
+        print(f"  • Chrome: {data.get('chrome', 'Unknown')}")
+        print(f"  • Languages: {data.get('languages', 'Unknown')}")
+        print(f"  • Plugins: {data.get('pluginsLength', 'Unknown')}")
+        print(f"  • User Agent: {data.get('userAgent', 'Unknown')[:60]}...")
+    
+    # Test 2: Cloudflare Site
+    print(f"\n\n{Fore.YELLOW}2. Cloudflare Protected Site Test{Style.RESET_ALL}")
+    print("-" * 40)
+    
+    # Without stealth
+    regular_cf = await test_cloudflare_site(use_stealth=False)
+    print(f"{Fore.RED}Without Stealth:{Style.RESET_ALL}")
+    print(f"  • Success: {regular_cf['success']}")
+    print(f"  • Cloudflare Challenge: {regular_cf['cloudflare_challenge']}")
+    print(f"  • Status Code: {regular_cf['status_code']}")
+    print(f"  • Page Title: {regular_cf['page_title']}")
+    
+    # With stealth
+    stealth_cf = await test_cloudflare_site(use_stealth=True)
+    print(f"\n{Fore.GREEN}With Stealth:{Style.RESET_ALL}")
+    print(f"  • Success: {stealth_cf['success']}")
+    print(f"  • Cloudflare Challenge: {stealth_cf['cloudflare_challenge']}")
+    print(f"  • Status Code: {stealth_cf['status_code']}")
+    print(f"  • Page Title: {stealth_cf['page_title']}")
+    
+    # Test 3: Anti-bot Site
+    print(f"\n\n{Fore.YELLOW}3. Anti-Bot Site Test{Style.RESET_ALL}")
+    print("-" * 40)
+    
+    # Without stealth
+    regular_antibot = await test_anti_bot_site(use_stealth=False)
+    print(f"{Fore.RED}Without Stealth:{Style.RESET_ALL}")
+    print(f"  • Success: {regular_antibot['success']}")
+    print(f"  • Blocked: {regular_antibot['blocked']}")
+    print(f"  • Status Code: {regular_antibot['status_code']}")
+    print(f"  • Page Title: {regular_antibot['page_title']}")
+    
+    # With stealth
+    stealth_antibot = await test_anti_bot_site(use_stealth=True)
+    print(f"\n{Fore.GREEN}With Stealth:{Style.RESET_ALL}")
+    print(f"  • Success: {stealth_antibot['success']}")
+    print(f"  • Blocked: {stealth_antibot['blocked']}")
+    print(f"  • Status Code: {stealth_antibot['status_code']}")
+    print(f"  • Page Title: {stealth_antibot['page_title']}")
+    
+    # Summary
+    print(f"\n{Fore.CYAN}{'='*60}{Style.RESET_ALL}")
+    print(f"{Fore.CYAN}Summary:{Style.RESET_ALL}")
+    print(f"{Fore.CYAN}{'='*60}{Style.RESET_ALL}")
+    print(f"\nStealth mode helps bypass basic bot detection by:")
+    print(f"  • Hiding webdriver property")
+    print(f"  • Modifying browser fingerprints")
+    print(f"  • Adjusting navigator properties")
+    print(f"  • Emulating real browser plugin behavior")
+    print(f"\n{Fore.YELLOW}Note:{Style.RESET_ALL} Stealth mode is not a silver bullet.")
+    print(f"Advanced anti-bot systems may still detect automation.")
+    print(f"Always respect robots.txt and website terms of service.")
+
+
+async def stealth_best_practices():
+    """Demonstrate best practices for using stealth mode"""
+    
+    print(f"\n\n{Fore.CYAN}{'='*60}{Style.RESET_ALL}")
+    print(f"{Fore.CYAN}Stealth Mode Best Practices{Style.RESET_ALL}")
+    print(f"{Fore.CYAN}{'='*60}{Style.RESET_ALL}\n")
+    
+    # Best Practice 1: Combine with realistic behavior
+    print(f"{Fore.YELLOW}1. Combine with Realistic Behavior:{Style.RESET_ALL}")
+    
+    browser_config = BrowserConfig(
+        headless=False,
+        enable_stealth=True,
+        viewport_width=1920,
+        viewport_height=1080
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        # Simulate human-like behavior
+        human_behavior_script = """
+        (async () => {
+            // Wait random time between actions
+            const randomWait = () => Math.random() * 2000 + 1000;
+            
+            // Simulate reading
+            await new Promise(resolve => setTimeout(resolve, randomWait()));
+            
+            // Smooth scroll
+            const smoothScroll = async () => {
+                const totalHeight = document.body.scrollHeight;
+                const viewHeight = window.innerHeight;
+                let currentPosition = 0;
+                
+                while (currentPosition < totalHeight - viewHeight) {
+                    const scrollAmount = Math.random() * 300 + 100;
+                    window.scrollBy({
+                        top: scrollAmount,
+                        behavior: 'smooth'
+                    });
+                    currentPosition += scrollAmount;
+                    await new Promise(resolve => setTimeout(resolve, randomWait()));
+                }
+            };
+            
+            await smoothScroll();
+            console.log('Human-like behavior simulation completed');
+            return true;
+        })()
+        """
+        
+        config = CrawlerRunConfig(
+            js_code=human_behavior_script,
+            wait_until="networkidle",
+            delay_before_return_html=3.0,
+            capture_console_messages=True
+        )
+        
+        result = await crawler.arun(
+            url="https://example.com",
+            config=config
+        )
+        
+        print(f"  ✓ Simulated human-like scrolling and reading patterns")
+        print(f"  ✓ Added random delays between actions")
+        print(f"  ✓ Result: {result.success}")
+    
+    # Best Practice 2: Use appropriate viewport and user agent
+    print(f"\n{Fore.YELLOW}2. Use Realistic Viewport and User Agent:{Style.RESET_ALL}")
+    
+    # Get a realistic user agent
+    from crawl4ai.user_agent_generator import UserAgentGenerator
+    ua_generator = UserAgentGenerator()
+    
+    browser_config = BrowserConfig(
+        headless=True,
+        enable_stealth=True,
+        viewport_width=1920,
+        viewport_height=1080,
+        user_agent=ua_generator.generate(device_type="desktop", browser_type="chrome")
+    )
+    
+    print(f"  ✓ Using realistic viewport: 1920x1080")
+    print(f"  ✓ Using current Chrome user agent")
+    print(f"  ✓ Stealth mode will ensure consistency")
+    
+    # Best Practice 3: Manage request rate
+    print(f"\n{Fore.YELLOW}3. Manage Request Rate:{Style.RESET_ALL}")
+    print(f"  ✓ Add delays between requests")
+    print(f"  ✓ Randomize timing patterns")
+    print(f"  ✓ Respect robots.txt")
+    
+    # Best Practice 4: Session management
+    print(f"\n{Fore.YELLOW}4. Use Session Management:{Style.RESET_ALL}")
+    
+    browser_config = BrowserConfig(
+        headless=False,
+        enable_stealth=True
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        # Create a session for multiple requests
+        session_id = "stealth_session_1"
+        
+        config = CrawlerRunConfig(
+            session_id=session_id,
+            wait_until="domcontentloaded"
+        )
+        
+        # First request
+        result1 = await crawler.arun(
+            url="https://example.com",
+            config=config
+        )
+        
+        # Subsequent request reuses the same browser context
+        result2 = await crawler.arun(
+            url="https://example.com/about",
+            config=config
+        )
+        
+        print(f"  ✓ Reused browser session for multiple requests")
+        print(f"  ✓ Maintains cookies and state between requests")
+        print(f"  ✓ More efficient and realistic browsing pattern")
+    
+    print(f"\n{Fore.CYAN}{'='*60}{Style.RESET_ALL}")
+
+
+async def main():
+    """Run all examples"""
+    
+    # Run comparison tests
+    await compare_results()
+    
+    # Show best practices
+    await stealth_best_practices()
+    
+    print(f"\n{Fore.GREEN}Examples completed!{Style.RESET_ALL}")
+    print(f"\n{Fore.YELLOW}Remember:{Style.RESET_ALL}")
+    print(f"• Stealth mode helps with basic bot detection")
+    print(f"• Always respect website terms of service")
+    print(f"• Consider rate limiting and ethical scraping practices")
+    print(f"• For advanced protection, consider additional measures")
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

docs/examples/stealth_mode_quick_start.py

@@ -0,0 +1,215 @@
+"""
+Quick Start: Using Stealth Mode in Crawl4AI
+
+This example shows practical use cases for the stealth mode feature.
+Stealth mode helps bypass basic bot detection mechanisms.
+"""
+
+import asyncio
+from crawl4ai import AsyncWebCrawler, BrowserConfig, CrawlerRunConfig
+
+
+async def example_1_basic_stealth():
+    """Example 1: Basic stealth mode usage"""
+    print("\n=== Example 1: Basic Stealth Mode ===")
+    
+    # Enable stealth mode in browser config
+    browser_config = BrowserConfig(
+        enable_stealth=True,  # This is the key parameter
+        headless=True
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        result = await crawler.arun(url="https://example.com")
+        print(f"✓ Crawled {result.url} successfully")
+        print(f"✓ Title: {result.metadata.get('title', 'N/A')}")
+
+
+async def example_2_stealth_with_screenshot():
+    """Example 2: Stealth mode with screenshot to show detection results"""
+    print("\n=== Example 2: Stealth Mode Visual Verification ===")
+    
+    browser_config = BrowserConfig(
+        enable_stealth=True,
+        headless=False  # Set to False to see the browser
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        config = CrawlerRunConfig(
+            screenshot=True,
+            wait_until="networkidle"
+        )
+        
+        result = await crawler.arun(
+            url="https://bot.sannysoft.com",
+            config=config
+        )
+        
+        if result.success:
+            print(f"✓ Successfully crawled bot detection site")
+            print(f"✓ With stealth enabled, many detection tests should show as passed")
+            
+            if result.screenshot:
+                # Save screenshot for verification
+                import base64
+                with open("stealth_detection_results.png", "wb") as f:
+                    f.write(base64.b64decode(result.screenshot))
+                print(f"✓ Screenshot saved as 'stealth_detection_results.png'")
+                print(f"  Check the screenshot to see detection results!")
+
+
+async def example_3_stealth_for_protected_sites():
+    """Example 3: Using stealth for sites with bot protection"""
+    print("\n=== Example 3: Stealth for Protected Sites ===")
+    
+    browser_config = BrowserConfig(
+        enable_stealth=True,
+        headless=True,
+        viewport_width=1920,
+        viewport_height=1080
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        # Add human-like behavior
+        config = CrawlerRunConfig(
+            wait_until="networkidle",
+            delay_before_return_html=2.0,  # Wait 2 seconds
+            js_code="""
+            // Simulate human-like scrolling
+            window.scrollTo({
+                top: document.body.scrollHeight / 2,
+                behavior: 'smooth'
+            });
+            """
+        )
+        
+        # Try accessing a site that might have bot protection
+        result = await crawler.arun(
+            url="https://www.g2.com/products/slack/reviews",
+            config=config
+        )
+        
+        if result.success:
+            print(f"✓ Successfully accessed protected site")
+            print(f"✓ Retrieved {len(result.html)} characters of HTML")
+        else:
+            print(f"✗ Failed to access site: {result.error_message}")
+
+
+async def example_4_stealth_with_sessions():
+    """Example 4: Stealth mode with session management"""
+    print("\n=== Example 4: Stealth + Session Management ===")
+    
+    browser_config = BrowserConfig(
+        enable_stealth=True,
+        headless=False
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        session_id = "my_stealth_session"
+        
+        # First request - establish session
+        config = CrawlerRunConfig(
+            session_id=session_id,
+            wait_until="domcontentloaded"
+        )
+        
+        result1 = await crawler.arun(
+            url="https://news.ycombinator.com",
+            config=config
+        )
+        print(f"✓ First request completed: {result1.url}")
+        
+        # Second request - reuse session
+        await asyncio.sleep(2)  # Brief delay between requests
+        
+        result2 = await crawler.arun(
+            url="https://news.ycombinator.com/best",
+            config=config
+        )
+        print(f"✓ Second request completed: {result2.url}")
+        print(f"✓ Session reused, maintaining cookies and state")
+
+
+async def example_5_stealth_comparison():
+    """Example 5: Compare results with and without stealth using screenshots"""
+    print("\n=== Example 5: Stealth Mode Comparison ===")
+    
+    test_url = "https://bot.sannysoft.com"
+    
+    # First test WITHOUT stealth
+    print("\nWithout stealth:")
+    regular_config = BrowserConfig(
+        enable_stealth=False,
+        headless=True
+    )
+    
+    async with AsyncWebCrawler(config=regular_config) as crawler:
+        config = CrawlerRunConfig(
+            screenshot=True,
+            wait_until="networkidle"
+        )
+        result = await crawler.arun(url=test_url, config=config)
+        
+        if result.success and result.screenshot:
+            import base64
+            with open("comparison_without_stealth.png", "wb") as f:
+                f.write(base64.b64decode(result.screenshot))
+            print(f"  ✓ Screenshot saved: comparison_without_stealth.png")
+            print(f"  Many tests will show as FAILED (red)")
+    
+    # Then test WITH stealth
+    print("\nWith stealth:")
+    stealth_config = BrowserConfig(
+        enable_stealth=True,
+        headless=True
+    )
+    
+    async with AsyncWebCrawler(config=stealth_config) as crawler:
+        config = CrawlerRunConfig(
+            screenshot=True,
+            wait_until="networkidle"
+        )
+        result = await crawler.arun(url=test_url, config=config)
+        
+        if result.success and result.screenshot:
+            import base64
+            with open("comparison_with_stealth.png", "wb") as f:
+                f.write(base64.b64decode(result.screenshot))
+            print(f"  ✓ Screenshot saved: comparison_with_stealth.png")
+            print(f"  More tests should show as PASSED (green)")
+    
+    print("\nCompare the two screenshots to see the difference!")
+
+
+async def main():
+    """Run all examples"""
+    print("Crawl4AI Stealth Mode Examples")
+    print("==============================")
+    
+    # Run basic example
+    await example_1_basic_stealth()
+    
+    # Run screenshot verification example
+    await example_2_stealth_with_screenshot()
+    
+    # Run protected site example
+    await example_3_stealth_for_protected_sites()
+    
+    # Run session example
+    await example_4_stealth_with_sessions()
+    
+    # Run comparison example
+    await example_5_stealth_comparison()
+    
+    print("\n" + "="*50)
+    print("Tips for using stealth mode effectively:")
+    print("- Use realistic viewport sizes (1920x1080, 1366x768)")
+    print("- Add delays between requests to appear more human")
+    print("- Combine with session management for better results")
+    print("- Remember: stealth mode is for legitimate scraping only")
+    print("="*50)
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

docs/examples/stealth_test_simple.py

@@ -0,0 +1,62 @@
+"""
+Simple test to verify stealth mode is working
+"""
+
+import asyncio
+from crawl4ai import AsyncWebCrawler, BrowserConfig, CrawlerRunConfig
+
+
+async def test_stealth():
+    """Test stealth mode effectiveness"""
+    
+    # Test WITHOUT stealth
+    print("=== WITHOUT Stealth ===")
+    config1 = BrowserConfig(
+        headless=False,
+        enable_stealth=False
+    )
+    
+    async with AsyncWebCrawler(config=config1) as crawler:
+        result = await crawler.arun(
+            url="https://bot.sannysoft.com",
+            config=CrawlerRunConfig(
+                wait_until="networkidle",
+                screenshot=True
+            )
+        )
+        print(f"Success: {result.success}")
+        # Take screenshot
+        if result.screenshot:
+            with open("without_stealth.png", "wb") as f:
+                import base64
+                f.write(base64.b64decode(result.screenshot))
+            print("Screenshot saved: without_stealth.png")
+    
+    # Test WITH stealth
+    print("\n=== WITH Stealth ===")
+    config2 = BrowserConfig(
+        headless=False,
+        enable_stealth=True
+    )
+    
+    async with AsyncWebCrawler(config=config2) as crawler:
+        result = await crawler.arun(
+            url="https://bot.sannysoft.com",
+            config=CrawlerRunConfig(
+                wait_until="networkidle",
+                screenshot=True
+            )
+        )
+        print(f"Success: {result.success}")
+        # Take screenshot
+        if result.screenshot:
+            with open("with_stealth.png", "wb") as f:
+                import base64
+                f.write(base64.b64decode(result.screenshot))
+            print("Screenshot saved: with_stealth.png")
+    
+    print("\nCheck the screenshots to see the difference in bot detection results!")
+
+
+if __name__ == "__main__":
+    asyncio.run(test_stealth())

docs/examples/undetectability/undetected_basic_test.py

@@ -0,0 +1,74 @@
+"""
+Basic Undetected Browser Test
+Simple example to test if undetected mode works
+"""
+
+import asyncio
+from crawl4ai import AsyncWebCrawler, BrowserConfig
+
+async def test_regular_mode():
+    """Test with regular browser"""
+    print("Testing Regular Browser Mode...")
+    browser_config = BrowserConfig(
+        headless=False,
+        verbose=True
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        result = await crawler.arun(url="https://www.example.com")
+        print(f"Regular Mode - Success: {result.success}")
+        print(f"Regular Mode - Status: {result.status_code}")
+        print(f"Regular Mode - Content length: {len(result.markdown.raw_markdown)}")
+        print(f"Regular Mode - First 100 chars: {result.markdown.raw_markdown[:100]}...")
+        return result.success
+
+async def test_undetected_mode():
+    """Test with undetected browser"""
+    print("\nTesting Undetected Browser Mode...")
+    from crawl4ai import UndetectedAdapter
+    from crawl4ai.async_crawler_strategy import AsyncPlaywrightCrawlerStrategy
+    
+    browser_config = BrowserConfig(
+        headless=False,
+        verbose=True
+    )
+    
+    # Create undetected adapter
+    undetected_adapter = UndetectedAdapter()
+    
+    # Create strategy with undetected adapter
+    crawler_strategy = AsyncPlaywrightCrawlerStrategy(
+        browser_config=browser_config,
+        browser_adapter=undetected_adapter
+    )
+    
+    async with AsyncWebCrawler(
+        crawler_strategy=crawler_strategy,
+        config=browser_config
+    ) as crawler:
+        result = await crawler.arun(url="https://www.example.com")
+        print(f"Undetected Mode - Success: {result.success}")
+        print(f"Undetected Mode - Status: {result.status_code}")
+        print(f"Undetected Mode - Content length: {len(result.markdown.raw_markdown)}")
+        print(f"Undetected Mode - First 100 chars: {result.markdown.raw_markdown[:100]}...")
+        return result.success
+
+async def main():
+    """Run both tests"""
+    print("🤖 Crawl4AI Basic Adapter Test\n")
+    
+    # Test regular mode
+    regular_success = await test_regular_mode()
+    
+    # Test undetected mode
+    undetected_success = await test_undetected_mode()
+    
+    # Summary
+    print("\n" + "="*50)
+    print("Summary:")
+    print(f"Regular Mode: {'✅ Success' if regular_success else '❌ Failed'}")
+    print(f"Undetected Mode: {'✅ Success' if undetected_success else '❌ Failed'}")
+    print("="*50)
+
+if __name__ == "__main__":
+    asyncio.run(main())

docs/examples/undetectability/undetected_bot_test.py

@@ -0,0 +1,155 @@
+"""
+Bot Detection Test - Compare Regular vs Undetected
+Tests browser fingerprinting differences at bot.sannysoft.com
+"""
+
+import asyncio
+from crawl4ai import (
+    AsyncWebCrawler, 
+    BrowserConfig, 
+    CrawlerRunConfig,
+    UndetectedAdapter,
+    CrawlResult
+)
+from crawl4ai.async_crawler_strategy import AsyncPlaywrightCrawlerStrategy
+
+# Bot detection test site
+TEST_URL = "https://bot.sannysoft.com"
+
+def analyze_bot_detection(result: CrawlResult) -> dict:
+    """Analyze bot detection results from the page"""
+    detections = {
+        "webdriver": False,
+        "headless": False, 
+        "automation": False,
+        "user_agent": False,
+        "total_tests": 0,
+        "failed_tests": 0
+    }
+    
+    if not result.success or not result.html:
+        return detections
+    
+    # Look for specific test results in the HTML
+    html_lower = result.html.lower()
+    
+    # Check for common bot indicators
+    if "webdriver" in html_lower and ("fail" in html_lower or "true" in html_lower):
+        detections["webdriver"] = True
+        detections["failed_tests"] += 1
+    
+    if "headless" in html_lower and ("fail" in html_lower or "true" in html_lower):
+        detections["headless"] = True
+        detections["failed_tests"] += 1
+    
+    if "automation" in html_lower and "detected" in html_lower:
+        detections["automation"] = True
+        detections["failed_tests"] += 1
+    
+    # Count total tests (approximate)
+    detections["total_tests"] = html_lower.count("test") + html_lower.count("check")
+    
+    return detections
+
+async def test_browser_mode(adapter_name: str, adapter=None):
+    """Test a browser mode and return results"""
+    print(f"\n{'='*60}")
+    print(f"Testing: {adapter_name}")
+    print(f"{'='*60}")
+    
+    browser_config = BrowserConfig(
+        headless=False,  # Run in headed mode for better results
+        verbose=True,
+        viewport_width=1920,
+        viewport_height=1080,
+    )
+    
+    if adapter:
+        # Use undetected mode
+        crawler_strategy = AsyncPlaywrightCrawlerStrategy(
+            browser_config=browser_config,
+            browser_adapter=adapter
+        )
+        crawler = AsyncWebCrawler(
+            crawler_strategy=crawler_strategy,
+            config=browser_config
+        )
+    else:
+        # Use regular mode
+        crawler = AsyncWebCrawler(config=browser_config)
+    
+    async with crawler:
+        config = CrawlerRunConfig(
+            delay_before_return_html=3.0,  # Let detection scripts run
+            wait_for_images=True,
+            screenshot=True,
+            simulate_user=False,  # Don't simulate for accurate detection
+        )
+        
+        result = await crawler.arun(url=TEST_URL, config=config)
+        
+        print(f"\n✓ Success: {result.success}")
+        print(f"✓ Status Code: {result.status_code}")
+        
+        if result.success:
+            # Analyze detection results
+            detections = analyze_bot_detection(result)
+            
+            print(f"\n🔍 Bot Detection Analysis:")
+            print(f"  - WebDriver Detected: {'❌ Yes' if detections['webdriver'] else '✅ No'}")
+            print(f"  - Headless Detected: {'❌ Yes' if detections['headless'] else '✅ No'}")
+            print(f"  - Automation Detected: {'❌ Yes' if detections['automation'] else '✅ No'}")
+            print(f"  - Failed Tests: {detections['failed_tests']}")
+            
+            # Show some content
+            if result.markdown.raw_markdown:
+                print(f"\nContent preview:")
+                lines = result.markdown.raw_markdown.split('\n')
+                for line in lines[:20]:  # Show first 20 lines
+                    if any(keyword in line.lower() for keyword in ['test', 'pass', 'fail', 'yes', 'no']):
+                        print(f"  {line.strip()}")
+        
+        return result, detections if result.success else {}
+
+async def main():
+    """Run the comparison"""
+    print("🤖 Crawl4AI - Bot Detection Test")
+    print(f"Testing at: {TEST_URL}")
+    print("This site runs various browser fingerprinting tests\n")
+    
+    # Test regular browser
+    regular_result, regular_detections = await test_browser_mode("Regular Browser")
+    
+    # Small delay
+    await asyncio.sleep(2)
+    
+    # Test undetected browser
+    undetected_adapter = UndetectedAdapter()
+    undetected_result, undetected_detections = await test_browser_mode(
+        "Undetected Browser", 
+        undetected_adapter
+    )
+    
+    # Summary comparison
+    print(f"\n{'='*60}")
+    print("COMPARISON SUMMARY")
+    print(f"{'='*60}")
+    
+    print(f"\n{'Test':<25} {'Regular':<15} {'Undetected':<15}")
+    print(f"{'-'*55}")
+    
+    if regular_detections and undetected_detections:
+        print(f"{'WebDriver Detection':<25} {'❌ Detected' if regular_detections['webdriver'] else '✅ Passed':<15} {'❌ Detected' if undetected_detections['webdriver'] else '✅ Passed':<15}")
+        print(f"{'Headless Detection':<25} {'❌ Detected' if regular_detections['headless'] else '✅ Passed':<15} {'❌ Detected' if undetected_detections['headless'] else '✅ Passed':<15}")
+        print(f"{'Automation Detection':<25} {'❌ Detected' if regular_detections['automation'] else '✅ Passed':<15} {'❌ Detected' if undetected_detections['automation'] else '✅ Passed':<15}")
+        print(f"{'Failed Tests':<25} {regular_detections['failed_tests']:<15} {undetected_detections['failed_tests']:<15}")
+    
+    print(f"\n{'='*60}")
+    
+    if undetected_detections.get('failed_tests', 0) < regular_detections.get('failed_tests', 1):
+        print("✅ Undetected browser performed better at evading detection!")
+    else:
+        print("ℹ️  Both browsers had similar detection results")
+
+if __name__ == "__main__":
+    asyncio.run(main())

docs/examples/undetectability/undetected_cloudflare_test.py

@@ -0,0 +1,164 @@
+"""
+Undetected Browser Test - Cloudflare Protected Site
+Tests the difference between regular and undetected modes on a Cloudflare-protected site
+"""
+
+import asyncio
+from crawl4ai import (
+    AsyncWebCrawler, 
+    BrowserConfig, 
+    CrawlerRunConfig,
+    UndetectedAdapter
+)
+from crawl4ai.async_crawler_strategy import AsyncPlaywrightCrawlerStrategy
+
+# Test URL with Cloudflare protection
+TEST_URL = "https://nowsecure.nl"
+
+async def test_regular_browser():
+    """Test with regular browser - likely to be blocked"""
+    print("=" * 60)
+    print("Testing with Regular Browser")
+    print("=" * 60)
+    
+    browser_config = BrowserConfig(
+        headless=False,
+        verbose=True,
+        viewport_width=1920,
+        viewport_height=1080,
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        config = CrawlerRunConfig(
+            delay_before_return_html=2.0,
+            simulate_user=True,
+            magic=True,  # Try with magic mode too
+        )
+        
+        result = await crawler.arun(url=TEST_URL, config=config)
+        
+        print(f"\n✓ Success: {result.success}")
+        print(f"✓ Status Code: {result.status_code}")
+        print(f"✓ HTML Length: {len(result.html)}")
+        
+        # Check for Cloudflare challenge
+        if result.html:
+            cf_indicators = [
+                "Checking your browser",
+                "Please stand by",
+                "cloudflare",
+                "cf-browser-verification",
+                "Access denied",
+                "Ray ID"
+            ]
+            
+            detected = False
+            for indicator in cf_indicators:
+                if indicator.lower() in result.html.lower():
+                    print(f"⚠️  Cloudflare Challenge Detected: '{indicator}' found")
+                    detected = True
+                    break
+            
+            if not detected and len(result.markdown.raw_markdown) > 100:
+                print("✅ Successfully bypassed Cloudflare!")
+                print(f"Content preview: {result.markdown.raw_markdown[:200]}...")
+            elif not detected:
+                print("⚠️  Page loaded but content seems minimal")
+        
+        return result
+
+async def test_undetected_browser():
+    """Test with undetected browser - should bypass Cloudflare"""
+    print("\n" + "=" * 60)
+    print("Testing with Undetected Browser")
+    print("=" * 60)
+    
+    browser_config = BrowserConfig(
+        headless=False,  # Headless is easier to detect
+        verbose=True,
+        viewport_width=1920,
+        viewport_height=1080,
+    )
+    
+    # Create undetected adapter
+    undetected_adapter = UndetectedAdapter()
+    
+    # Create strategy with undetected adapter
+    crawler_strategy = AsyncPlaywrightCrawlerStrategy(
+        browser_config=browser_config,
+        browser_adapter=undetected_adapter
+    )
+    
+    async with AsyncWebCrawler(
+        crawler_strategy=crawler_strategy,
+        config=browser_config
+    ) as crawler:
+        config = CrawlerRunConfig(
+            delay_before_return_html=2.0,
+            simulate_user=True,
+        )
+        
+        result = await crawler.arun(url=TEST_URL, config=config)
+        
+        print(f"\n✓ Success: {result.success}")
+        print(f"✓ Status Code: {result.status_code}")
+        print(f"✓ HTML Length: {len(result.html)}")
+        
+        # Check for Cloudflare challenge
+        if result.html:
+            cf_indicators = [
+                "Checking your browser",
+                "Please stand by",
+                "cloudflare",
+                "cf-browser-verification",
+                "Access denied",
+                "Ray ID"
+            ]
+            
+            detected = False
+            for indicator in cf_indicators:
+                if indicator.lower() in result.html.lower():
+                    print(f"⚠️  Cloudflare Challenge Detected: '{indicator}' found")
+                    detected = True
+                    break
+            
+            if not detected and len(result.markdown.raw_markdown) > 100:
+                print("✅ Successfully bypassed Cloudflare!")
+                print(f"Content preview: {result.markdown.raw_markdown[:200]}...")
+            elif not detected:
+                print("⚠️  Page loaded but content seems minimal")
+        
+        return result
+
+async def main():
+    """Compare regular vs undetected browser"""
+    print("🤖 Crawl4AI - Cloudflare Bypass Test")
+    print(f"Testing URL: {TEST_URL}\n")
+    
+    # Test regular browser
+    regular_result = await test_regular_browser()
+    
+    # Small delay
+    await asyncio.sleep(2)
+    
+    # Test undetected browser
+    undetected_result = await test_undetected_browser()
+    
+    # Summary
+    print("\n" + "=" * 60)
+    print("SUMMARY")
+    print("=" * 60)
+    print(f"Regular Browser:")
+    print(f"  - Success: {regular_result.success}")
+    print(f"  - Content Length: {len(regular_result.markdown.raw_markdown) if regular_result.markdown else 0}")
+    
+    print(f"\nUndetected Browser:")
+    print(f"  - Success: {undetected_result.success}")
+    print(f"  - Content Length: {len(undetected_result.markdown.raw_markdown) if undetected_result.markdown else 0}")
+    
+    if undetected_result.success and len(undetected_result.markdown.raw_markdown) > len(regular_result.markdown.raw_markdown):
+        print("\n✅ Undetected browser successfully bypassed protection!")
+    print("=" * 60)
+
+if __name__ == "__main__":
+    asyncio.run(main())

docs/examples/undetectability/undetected_vs_regular_comparison.py

@@ -0,0 +1,184 @@
+"""
+Undetected vs Regular Browser Comparison
+This example demonstrates the difference between regular and undetected browser modes
+when accessing sites with bot detection services.
+
+Based on tested anti-bot services:
+- Cloudflare
+- Kasada
+- Akamai
+- DataDome
+- Bet365
+- And others
+"""
+
+import asyncio
+from crawl4ai import (
+    AsyncWebCrawler,
+    BrowserConfig,
+    CrawlerRunConfig,
+    PlaywrightAdapter,
+    UndetectedAdapter,
+    CrawlResult
+)
+from crawl4ai.async_crawler_strategy import AsyncPlaywrightCrawlerStrategy
+
+
+# Test URLs for various bot detection services
+TEST_SITES = {
+    "Cloudflare Protected": "https://nowsecure.nl",
+    # "Bot Detection Test": "https://bot.sannysoft.com",
+    # "Fingerprint Test": "https://fingerprint.com/products/bot-detection",
+    # "Browser Scan": "https://browserscan.net",
+    # "CreepJS": "https://abrahamjuliot.github.io/creepjs",
+}
+
+
+async def test_with_adapter(url: str, adapter_name: str, adapter):
+    """Test a URL with a specific adapter"""
+    browser_config = BrowserConfig(
+        headless=False,  # Better for avoiding detection
+        viewport_width=1920,
+        viewport_height=1080,
+        verbose=True,
+    )
+    
+    # Create the crawler strategy with the adapter
+    crawler_strategy = AsyncPlaywrightCrawlerStrategy(
+        browser_config=browser_config,
+        browser_adapter=adapter
+    )
+    
+    print(f"\n{'='*60}")
+    print(f"Testing with {adapter_name} adapter")
+    print(f"URL: {url}")
+    print(f"{'='*60}")
+    
+    try:
+        async with AsyncWebCrawler(
+            crawler_strategy=crawler_strategy,
+            config=browser_config
+        ) as crawler:
+            crawler_config = CrawlerRunConfig(
+                delay_before_return_html=3.0,  # Give page time to load
+                wait_for_images=True,
+                screenshot=True,
+                simulate_user=True,  # Add user simulation
+            )
+            
+            result: CrawlResult = await crawler.arun(
+                url=url,
+                config=crawler_config
+            )
+            
+            # Check results
+            print(f"✓ Status Code: {result.status_code}")
+            print(f"✓ Success: {result.success}")
+            print(f"✓ HTML Length: {len(result.html)}")
+            print(f"✓ Markdown Length: {len(result.markdown.raw_markdown)}")
+            
+            # Check for common bot detection indicators
+            detection_indicators = [
+                "Access denied",
+                "Please verify you are human",
+                "Checking your browser",
+                "Enable JavaScript",
+                "captcha",
+                "403 Forbidden",
+                "Bot detection",
+                "Security check"
+            ]
+            
+            content_lower = result.markdown.raw_markdown.lower()
+            detected = False
+            for indicator in detection_indicators:
+                if indicator.lower() in content_lower:
+                    print(f"⚠️  Possible detection: Found '{indicator}'")
+                    detected = True
+                    break
+            
+            if not detected:
+                print("✅ No obvious bot detection triggered!")
+                # Show first 200 chars of content
+                print(f"Content preview: {result.markdown.raw_markdown[:200]}...")
+            
+            return result.success and not detected
+            
+    except Exception as e:
+        print(f"❌ Error: {str(e)}")
+        return False
+
+
+async def compare_adapters(url: str, site_name: str):
+    """Compare regular and undetected adapters on the same URL"""
+    print(f"\n{'#'*60}")
+    print(f"# Testing: {site_name}")
+    print(f"{'#'*60}")
+    
+    # Test with regular adapter
+    regular_adapter = PlaywrightAdapter()
+    regular_success = await test_with_adapter(url, "Regular", regular_adapter)
+    
+    # Small delay between tests
+    await asyncio.sleep(2)
+    
+    # Test with undetected adapter
+    undetected_adapter = UndetectedAdapter()
+    undetected_success = await test_with_adapter(url, "Undetected", undetected_adapter)
+    
+    # Summary
+    print(f"\n{'='*60}")
+    print(f"Summary for {site_name}:")
+    print(f"Regular Adapter: {'✅ Passed' if regular_success else '❌ Blocked/Detected'}")
+    print(f"Undetected Adapter: {'✅ Passed' if undetected_success else '❌ Blocked/Detected'}")
+    print(f"{'='*60}")
+    
+    return regular_success, undetected_success
+
+
+async def main():
+    """Run comparison tests on multiple sites"""
+    print("🤖 Crawl4AI Browser Adapter Comparison")
+    print("Testing regular vs undetected browser modes\n")
+    
+    results = {}
+    
+    # Test each site
+    for site_name, url in TEST_SITES.items():
+        regular, undetected = await compare_adapters(url, site_name)
+        results[site_name] = {
+            "regular": regular,
+            "undetected": undetected
+        }
+        
+        # Delay between different sites
+        await asyncio.sleep(3)
+    
+    # Final summary
+    print(f"\n{'#'*60}")
+    print("# FINAL RESULTS")
+    print(f"{'#'*60}")
+    print(f"{'Site':<30} {'Regular':<15} {'Undetected':<15}")
+    print(f"{'-'*60}")
+    
+    for site, result in results.items():
+        regular_status = "✅ Passed" if result["regular"] else "❌ Blocked"
+        undetected_status = "✅ Passed" if result["undetected"] else "❌ Blocked"
+        print(f"{site:<30} {regular_status:<15} {undetected_status:<15}")
+    
+    # Calculate success rates
+    regular_success = sum(1 for r in results.values() if r["regular"])
+    undetected_success = sum(1 for r in results.values() if r["undetected"])
+    total = len(results)
+    
+    print(f"\n{'='*60}")
+    print(f"Success Rates:")
+    print(f"Regular Adapter: {regular_success}/{total} ({regular_success/total*100:.1f}%)")
+    print(f"Undetected Adapter: {undetected_success}/{total} ({undetected_success/total*100:.1f}%)")
+    print(f"{'='*60}")
+
+
+if __name__ == "__main__":
+    # Note: This example may take a while to run as it tests multiple sites
+    # You can comment out sites in TEST_SITES to run faster tests
+    asyncio.run(main())

docs/examples/undetected_simple_demo.py

@@ -0,0 +1,118 @@
+"""
+Simple Undetected Browser Demo
+Demonstrates the basic usage of undetected browser mode
+"""
+
+import asyncio
+from crawl4ai import (
+    AsyncWebCrawler, 
+    BrowserConfig, 
+    CrawlerRunConfig,
+    UndetectedAdapter
+)
+from crawl4ai.async_crawler_strategy import AsyncPlaywrightCrawlerStrategy
+
+async def crawl_with_regular_browser(url: str):
+    """Crawl with regular browser"""
+    print("\n[Regular Browser Mode]")
+    browser_config = BrowserConfig(
+        headless=False,
+        verbose=True,
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        result = await crawler.arun(
+            url=url,
+            config=CrawlerRunConfig(
+                delay_before_return_html=2.0
+            )
+        )
+        
+        print(f"Success: {result.success}")
+        print(f"Status: {result.status_code}")
+        print(f"Content length: {len(result.markdown.raw_markdown)}")
+        
+        # Check for bot detection keywords
+        content = result.markdown.raw_markdown.lower()
+        if any(word in content for word in ["cloudflare", "checking your browser", "please wait"]):
+            print("⚠️  Bot detection triggered!")
+        else:
+            print("✅ Page loaded successfully")
+        
+        return result
+
+async def crawl_with_undetected_browser(url: str):
+    """Crawl with undetected browser"""
+    print("\n[Undetected Browser Mode]")
+    browser_config = BrowserConfig(
+        headless=False,
+        verbose=True,
+    )
+    
+    # Create undetected adapter and strategy
+    undetected_adapter = UndetectedAdapter()
+    crawler_strategy = AsyncPlaywrightCrawlerStrategy(
+        browser_config=browser_config,
+        browser_adapter=undetected_adapter
+    )
+    
+    async with AsyncWebCrawler(
+        crawler_strategy=crawler_strategy,
+        config=browser_config
+    ) as crawler:
+        result = await crawler.arun(
+            url=url,
+            config=CrawlerRunConfig(
+                delay_before_return_html=2.0
+            )
+        )
+        
+        print(f"Success: {result.success}")
+        print(f"Status: {result.status_code}")
+        print(f"Content length: {len(result.markdown.raw_markdown)}")
+        
+        # Check for bot detection keywords
+        content = result.markdown.raw_markdown.lower()
+        if any(word in content for word in ["cloudflare", "checking your browser", "please wait"]):
+            print("⚠️  Bot detection triggered!")
+        else:
+            print("✅ Page loaded successfully")
+        
+        return result
+
+async def main():
+    """Demo comparing regular vs undetected modes"""
+    print("🤖 Crawl4AI Undetected Browser Demo")
+    print("="*50)
+    
+    # Test URLs - you can change these
+    test_urls = [
+        "https://www.example.com",  # Simple site
+        "https://httpbin.org/headers",  # Shows request headers
+    ]
+    
+    for url in test_urls:
+        print(f"\n📍 Testing URL: {url}")
+        
+        # Test with regular browser
+        regular_result = await crawl_with_regular_browser(url)
+        
+        # Small delay
+        await asyncio.sleep(2)
+        
+        # Test with undetected browser
+        undetected_result = await crawl_with_undetected_browser(url)
+        
+        # Compare results
+        print(f"\n📊 Comparison for {url}:")
+        print(f"Regular browser content: {len(regular_result.markdown.raw_markdown)} chars")
+        print(f"Undetected browser content: {len(undetected_result.markdown.raw_markdown)} chars")
+        
+        if url == "https://httpbin.org/headers":
+            # Show headers for comparison
+            print("\nHeaders seen by server:")
+            print("Regular:", regular_result.markdown.raw_markdown[:500])
+            print("\nUndetected:", undetected_result.markdown.raw_markdown[:500])
+
+if __name__ == "__main__":
+    asyncio.run(main())

docs/md_v2/advanced/advanced-features.md

@@ -358,9 +358,77 @@ if __name__ == "__main__":
 
 ---
 
+---
+
+## 7. Anti-Bot Features (Stealth Mode & Undetected Browser)
+
+Crawl4AI provides two powerful features to bypass bot detection:
+
+### 7.1 Stealth Mode
+
+Stealth mode uses playwright-stealth to modify browser fingerprints and behaviors. Enable it with a simple flag:
+
+```python
+browser_config = BrowserConfig(
+    enable_stealth=True,  # Activates stealth mode
+    headless=False
+)
+```
+
+**When to use**: Sites with basic bot detection (checking navigator.webdriver, plugins, etc.)
+
+### 7.2 Undetected Browser
+
+For advanced bot detection, use the undetected browser adapter:
+
+```python
+from crawl4ai import UndetectedAdapter
+from crawl4ai.async_crawler_strategy import AsyncPlaywrightCrawlerStrategy
+
+# Create undetected adapter
+adapter = UndetectedAdapter()
+strategy = AsyncPlaywrightCrawlerStrategy(
+    browser_config=browser_config,
+    browser_adapter=adapter
+)
+
+async with AsyncWebCrawler(crawler_strategy=strategy, config=browser_config) as crawler:
+    # Your crawling code
+```
+
+**When to use**: Sites with sophisticated bot detection (Cloudflare, DataDome, etc.)
+
+### 7.3 Combining Both
+
+For maximum evasion, combine stealth mode with undetected browser:
+
+```python
+browser_config = BrowserConfig(
+    enable_stealth=True,  # Enable stealth
+    headless=False
+)
+
+adapter = UndetectedAdapter()  # Use undetected browser
+```
+
+### Choosing the Right Approach
+
+| Detection Level | Recommended Approach |
+|----------------|---------------------|
+| No protection | Regular browser |
+| Basic checks | Regular + Stealth mode |
+| Advanced protection | Undetected browser |
+| Maximum evasion | Undetected + Stealth mode |
+
+**Best Practice**: Start with regular browser + stealth mode. Only use undetected browser if needed, as it may be slightly slower.
+
+See [Undetected Browser Mode](undetected-browser.md) for detailed examples.
+
+---
+
 ## Conclusion & Next Steps
 
-You’ve now explored several **advanced** features:
+You've now explored several **advanced** features:
 
 - **Proxy Usage**  
 - **PDF & Screenshot** capturing for large or critical pages  
@@ -368,7 +436,10 @@ You’ve now explored several **advanced** features:
 - **Custom Headers** for language or specialized requests  
 - **Session Persistence** via storage state
 - **Robots.txt Compliance**
+- **Anti-Bot Features** (Stealth Mode & Undetected Browser)
 
-With these power tools, you can build robust scraping workflows that mimic real user behavior, handle secure sites, capture detailed snapshots, and manage sessions across multiple runs—streamlining your entire data collection pipeline.
+With these power tools, you can build robust scraping workflows that mimic real user behavior, handle secure sites, capture detailed snapshots, manage sessions across multiple runs, and bypass bot detection—streamlining your entire data collection pipeline.
 
-**Last Updated**: 2025-01-01
+**Note**: In future versions, we may enable stealth mode and undetected browser by default. For now, users should explicitly enable these features when needed.
+
+**Last Updated**: 2025-01-17

docs/md_v2/advanced/multi-url-crawling.md

@@ -404,7 +404,182 @@ for result in results:
         print(f"Duration: {dr.end_time - dr.start_time}")
 ```
 
-## 6. Summary
+## 6. URL-Specific Configurations
+
+When crawling diverse content types, you often need different configurations for different URLs. For example:
+- PDFs need specialized extraction
+- Blog pages benefit from content filtering
+- Dynamic sites need JavaScript execution
+- API endpoints need JSON parsing
+
+### 6.1 Basic URL Pattern Matching
+
+```python
+from crawl4ai import AsyncWebCrawler, CrawlerRunConfig, MatchMode
+from crawl4ai.processors.pdf import PDFContentScrapingStrategy
+from crawl4ai.extraction_strategy import JsonCssExtractionStrategy
+from crawl4ai.content_filter_strategy import PruningContentFilter
+from crawl4ai.markdown_generation_strategy import DefaultMarkdownGenerator
+
+async def crawl_mixed_content():
+    # Configure different strategies for different content
+    configs = [
+        # PDF files - specialized extraction
+        CrawlerRunConfig(
+            url_matcher="*.pdf",
+            scraping_strategy=PDFContentScrapingStrategy()
+        ),
+        
+        # Blog/article pages - content filtering
+        CrawlerRunConfig(
+            url_matcher=["*/blog/*", "*/article/*"],
+            markdown_generator=DefaultMarkdownGenerator(
+                content_filter=PruningContentFilter(threshold=0.48)
+            )
+        ),
+        
+        # Dynamic pages - JavaScript execution
+        CrawlerRunConfig(
+            url_matcher=lambda url: 'github.com' in url,
+            js_code="window.scrollTo(0, 500);"
+        ),
+        
+        # API endpoints - JSON extraction
+        CrawlerRunConfig(
+            url_matcher=lambda url: 'api' in url or url.endswith('.json'),
+            # Custome settings for JSON extraction
+        ),
+        
+        # Default config for everything else
+        CrawlerRunConfig()  # No url_matcher means it matches ALL URLs (fallback)
+    ]
+    
+    # Mixed URLs
+    urls = [
+        "https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf",
+        "https://blog.python.org/",
+        "https://github.com/microsoft/playwright",
+        "https://httpbin.org/json",
+        "https://example.com/"
+    ]
+    
+    async with AsyncWebCrawler() as crawler:
+        results = await crawler.arun_many(
+            urls=urls,
+            config=configs  # Pass list of configs
+        )
+        
+        for result in results:
+            print(f"{result.url}: {len(result.markdown)} chars")
+```
+
+### 6.2 Advanced Pattern Matching
+
+**Important**: A `CrawlerRunConfig` without `url_matcher` (or with `url_matcher=None`) matches ALL URLs. This makes it perfect as a default/fallback configuration.
+
+The `url_matcher` parameter supports three types of patterns:
+
+#### Glob Patterns (Strings)
+```python
+# Simple patterns
+"*.pdf"                    # Any PDF file
+"*/api/*"                  # Any URL with /api/ in path
+"https://*.example.com/*"  # Subdomain matching
+"*://example.com/blog/*"   # Any protocol
+```
+
+#### Custom Functions
+```python
+# Complex logic with lambdas
+lambda url: url.startswith('https://') and 'secure' in url
+lambda url: len(url) > 50 and url.count('/') > 5
+lambda url: any(domain in url for domain in ['api.', 'data.', 'feed.'])
+```
+
+#### Mixed Lists with AND/OR Logic
+```python
+# Combine multiple conditions
+CrawlerRunConfig(
+    url_matcher=[
+        "https://*",                        # Must be HTTPS
+        lambda url: 'internal' in url,      # Must contain 'internal'
+        lambda url: not url.endswith('.pdf') # Must not be PDF
+    ],
+    match_mode=MatchMode.AND  # ALL conditions must match
+)
+```
+
+### 6.3 Practical Example: News Site Crawler
+
+```python
+async def crawl_news_site():
+    dispatcher = MemoryAdaptiveDispatcher(
+        memory_threshold_percent=70.0,
+        rate_limiter=RateLimiter(base_delay=(1.0, 2.0))
+    )
+    
+    configs = [
+        # Homepage - light extraction
+        CrawlerRunConfig(
+            url_matcher=lambda url: url.rstrip('/') == 'https://news.ycombinator.com',
+            css_selector="nav, .headline",
+            extraction_strategy=None
+        ),
+        
+        # Article pages - full extraction
+        CrawlerRunConfig(
+            url_matcher="*/article/*",
+            extraction_strategy=CosineStrategy(
+                semantic_filter="article content",
+                word_count_threshold=100
+            ),
+            screenshot=True,
+            excluded_tags=["nav", "aside", "footer"]
+        ),
+        
+        # Author pages - metadata focus
+        CrawlerRunConfig(
+            url_matcher="*/author/*",
+            extraction_strategy=JsonCssExtractionStrategy({
+                "name": "h1.author-name",
+                "bio": ".author-bio",
+                "articles": "article.post-card h2"
+            })
+        ),
+        
+        # Everything else
+        CrawlerRunConfig()
+    ]
+    
+    async with AsyncWebCrawler() as crawler:
+        results = await crawler.arun_many(
+            urls=news_urls,
+            config=configs,
+            dispatcher=dispatcher
+        )
+```
+
+### 6.4 Best Practices
+
+1. **Order Matters**: Configs are evaluated in order - put specific patterns before general ones
+2. **Default Config Behavior**: 
+   - A config without `url_matcher` matches ALL URLs
+   - Always include a default config as the last item if you want to handle all URLs
+   - Without a default config, unmatched URLs will fail with "No matching configuration found"
+3. **Test Your Patterns**: Use the config's `is_match()` method to test patterns:
+   ```python
+   config = CrawlerRunConfig(url_matcher="*.pdf")
+   print(config.is_match("https://example.com/doc.pdf"))  # True
+   
+   default_config = CrawlerRunConfig()  # No url_matcher
+   print(default_config.is_match("https://any-url.com"))  # True - matches everything!
+   ```
+4. **Optimize for Performance**: 
+   - Disable JS for static content
+   - Skip screenshots for data APIs
+   - Use appropriate extraction strategies
+
+## 7. Summary
 
 1. **Two Dispatcher Types**:
 

docs/md_v2/advanced/undetected-browser.md

@@ -0,0 +1,394 @@
+# Undetected Browser Mode
+
+## Overview
+
+Crawl4AI offers two powerful anti-bot features to help you access websites with bot detection:
+
+1. **Stealth Mode** - Uses playwright-stealth to modify browser fingerprints and behaviors
+2. **Undetected Browser Mode** - Advanced browser adapter with deep-level patches for sophisticated bot detection
+
+This guide covers both features and helps you choose the right approach for your needs.
+
+## Anti-Bot Features Comparison
+
+| Feature | Regular Browser | Stealth Mode | Undetected Browser |
+|---------|----------------|--------------|-------------------|
+| WebDriver Detection | ❌ | ✅ | ✅ |
+| Navigator Properties | ❌ | ✅ | ✅ |
+| Plugin Emulation | ❌ | ✅ | ✅ |
+| CDP Detection | ❌ | Partial | ✅ |
+| Deep Browser Patches | ❌ | ❌ | ✅ |
+| Performance Impact | None | Minimal | Moderate |
+| Setup Complexity | None | None | Minimal |
+
+## When to Use Each Approach
+
+### Use Regular Browser + Stealth Mode When:
+- Sites have basic bot detection (checking navigator.webdriver, plugins, etc.)
+- You need good performance with basic protection
+- Sites check for common automation indicators
+
+### Use Undetected Browser When:
+- Sites employ sophisticated bot detection services (Cloudflare, DataDome, etc.)
+- Stealth mode alone isn't sufficient
+- You're willing to trade some performance for better evasion
+
+### Best Practice: Progressive Enhancement
+1. **Start with**: Regular browser + Stealth mode
+2. **If blocked**: Switch to Undetected browser
+3. **If still blocked**: Combine Undetected browser + Stealth mode
+
+## Stealth Mode
+
+Stealth mode is the simpler anti-bot solution that works with both regular and undetected browsers:
+
+```python
+from crawl4ai import AsyncWebCrawler, BrowserConfig
+
+# Enable stealth mode with regular browser
+browser_config = BrowserConfig(
+    enable_stealth=True,  # Simple flag to enable
+    headless=False       # Better for avoiding detection
+)
+
+async with AsyncWebCrawler(config=browser_config) as crawler:
+    result = await crawler.arun("https://example.com")
+```
+
+### What Stealth Mode Does:
+- Removes `navigator.webdriver` flag
+- Modifies browser fingerprints
+- Emulates realistic plugin behavior
+- Adjusts navigator properties
+- Fixes common automation leaks
+
+## Undetected Browser Mode
+
+For sites with sophisticated bot detection that stealth mode can't bypass, use the undetected browser adapter:
+
+### Key Features
+
+- **Drop-in Replacement**: Uses the same API as regular browser mode
+- **Enhanced Stealth**: Built-in patches to evade common detection methods
+- **Browser Adapter Pattern**: Seamlessly switch between regular and undetected modes
+- **Automatic Installation**: `crawl4ai-setup` installs all necessary browser dependencies
+
+### Quick Start
+
+```python
+import asyncio
+from crawl4ai import (
+    AsyncWebCrawler, 
+    BrowserConfig, 
+    CrawlerRunConfig,
+    UndetectedAdapter
+)
+from crawl4ai.async_crawler_strategy import AsyncPlaywrightCrawlerStrategy
+
+async def main():
+    # Create the undetected adapter
+    undetected_adapter = UndetectedAdapter()
+    
+    # Create browser config
+    browser_config = BrowserConfig(
+        headless=False,  # Headless mode can be detected easier
+        verbose=True,
+    )
+    
+    # Create the crawler strategy with undetected adapter
+    crawler_strategy = AsyncPlaywrightCrawlerStrategy(
+        browser_config=browser_config,
+        browser_adapter=undetected_adapter
+    )
+    
+    # Create the crawler with our custom strategy
+    async with AsyncWebCrawler(
+        crawler_strategy=crawler_strategy,
+        config=browser_config
+    ) as crawler:
+        # Your crawling code here
+        result = await crawler.arun(
+            url="https://example.com",
+            config=CrawlerRunConfig()
+        )
+        print(result.markdown[:500])
+
+asyncio.run(main())
+```
+
+## Combining Both Features
+
+For maximum evasion, combine stealth mode with undetected browser:
+
+```python
+from crawl4ai import AsyncWebCrawler, BrowserConfig, UndetectedAdapter
+from crawl4ai.async_crawler_strategy import AsyncPlaywrightCrawlerStrategy
+
+# Create browser config with stealth enabled
+browser_config = BrowserConfig(
+    enable_stealth=True,  # Enable stealth mode
+    headless=False
+)
+
+# Create undetected adapter
+adapter = UndetectedAdapter()
+
+# Create strategy with both features
+strategy = AsyncPlaywrightCrawlerStrategy(
+    browser_config=browser_config,
+    browser_adapter=adapter
+)
+
+async with AsyncWebCrawler(
+    crawler_strategy=strategy,
+    config=browser_config
+) as crawler:
+    result = await crawler.arun("https://protected-site.com")
+```
+
+## Examples
+
+### Example 1: Basic Stealth Mode
+
+```python
+import asyncio
+from crawl4ai import AsyncWebCrawler, BrowserConfig, CrawlerRunConfig
+
+async def test_stealth_mode():
+    # Simple stealth mode configuration
+    browser_config = BrowserConfig(
+        enable_stealth=True,
+        headless=False
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        result = await crawler.arun(
+            url="https://bot.sannysoft.com",
+            config=CrawlerRunConfig(screenshot=True)
+        )
+        
+        if result.success:
+            print("✓ Successfully accessed bot detection test site")
+            # Save screenshot to verify detection results
+            if result.screenshot:
+                import base64
+                with open("stealth_test.png", "wb") as f:
+                    f.write(base64.b64decode(result.screenshot))
+                print("✓ Screenshot saved - check for green (passed) tests")
+
+asyncio.run(test_stealth_mode())
+```
+
+### Example 2: Undetected Browser Mode
+
+```python
+import asyncio
+from crawl4ai import (
+    AsyncWebCrawler,
+    BrowserConfig,
+    CrawlerRunConfig,
+    UndetectedAdapter
+)
+from crawl4ai.async_crawler_strategy import AsyncPlaywrightCrawlerStrategy
+
+
+async def main():
+    # Create browser config
+    browser_config = BrowserConfig(
+        headless=False,
+        verbose=True,
+    )
+    
+    # Create the undetected adapter
+    undetected_adapter = UndetectedAdapter()
+    
+    # Create the crawler strategy with the undetected adapter
+    crawler_strategy = AsyncPlaywrightCrawlerStrategy(
+        browser_config=browser_config,
+        browser_adapter=undetected_adapter
+    )
+    
+    # Create the crawler with our custom strategy
+    async with AsyncWebCrawler(
+        crawler_strategy=crawler_strategy,
+        config=browser_config
+    ) as crawler:
+        # Configure the crawl
+        crawler_config = CrawlerRunConfig(
+            markdown_generator=DefaultMarkdownGenerator(
+                content_filter=PruningContentFilter()
+            ),
+            capture_console_messages=True,  # Test adapter console capture
+        )
+        
+        # Test on a site that typically detects bots
+        print("Testing undetected adapter...")
+        result: CrawlResult = await crawler.arun(
+            url="https://www.helloworld.org", 
+            config=crawler_config
+        )
+        
+        print(f"Status: {result.status_code}")
+        print(f"Success: {result.success}")
+        print(f"Console messages captured: {len(result.console_messages or [])}")
+        print(f"Markdown content (first 500 chars):\n{result.markdown.raw_markdown[:500]}")
+
+
+if __name__ == "__main__":
+    asyncio.run(main())
+```
+
+## Browser Adapter Pattern
+
+The undetected browser support is implemented using an adapter pattern, allowing seamless switching between different browser implementations:
+
+```python
+# Regular browser adapter (default)
+from crawl4ai import PlaywrightAdapter
+regular_adapter = PlaywrightAdapter()
+
+# Undetected browser adapter
+from crawl4ai import UndetectedAdapter
+undetected_adapter = UndetectedAdapter()
+```
+
+The adapter handles:
+- JavaScript execution
+- Console message capture
+- Error handling
+- Browser-specific optimizations
+
+## Best Practices
+
+1. **Avoid Headless Mode**: Detection is easier in headless mode
+   ```python
+   browser_config = BrowserConfig(headless=False)
+   ```
+
+2. **Use Reasonable Delays**: Don't rush through pages
+   ```python
+   crawler_config = CrawlerRunConfig(
+       wait_time=3.0,  # Wait 3 seconds after page load
+       delay_before_return_html=2.0  # Additional delay
+   )
+   ```
+
+3. **Rotate User Agents**: You can customize user agents
+   ```python
+   browser_config = BrowserConfig(
+       headers={"User-Agent": "your-user-agent"}
+   )
+   ```
+
+4. **Handle Failures Gracefully**: Some sites may still detect and block
+   ```python
+   if not result.success:
+       print(f"Crawl failed: {result.error_message}")
+   ```
+
+## Advanced Usage Tips
+
+### Progressive Detection Handling
+
+```python
+async def crawl_with_progressive_evasion(url):
+    # Step 1: Try regular browser with stealth
+    browser_config = BrowserConfig(
+        enable_stealth=True,
+        headless=False
+    )
+    
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        result = await crawler.arun(url)
+        if result.success and "Access Denied" not in result.html:
+            return result
+    
+    # Step 2: If blocked, try undetected browser
+    print("Regular + stealth blocked, trying undetected browser...")
+    
+    adapter = UndetectedAdapter()
+    strategy = AsyncPlaywrightCrawlerStrategy(
+        browser_config=browser_config,
+        browser_adapter=adapter
+    )
+    
+    async with AsyncWebCrawler(
+        crawler_strategy=strategy,
+        config=browser_config
+    ) as crawler:
+        result = await crawler.arun(url)
+        return result
+```
+
+## Installation
+
+The undetected browser dependencies are automatically installed when you run:
+
+```bash
+crawl4ai-setup
+```
+
+This command installs all necessary browser dependencies for both regular and undetected modes.
+
+## Limitations
+
+- **Performance**: Slightly slower than regular mode due to additional patches
+- **Headless Detection**: Some sites can still detect headless mode
+- **Resource Usage**: May use more resources than regular mode
+- **Not 100% Guaranteed**: Advanced anti-bot services are constantly evolving
+
+## Troubleshooting
+
+### Browser Not Found
+
+Run the setup command:
+```bash
+crawl4ai-setup
+```
+
+### Detection Still Occurring
+
+Try combining with other features:
+```python
+crawler_config = CrawlerRunConfig(
+    simulate_user=True,  # Add user simulation
+    magic=True,  # Enable magic mode
+    wait_time=5.0,  # Longer waits
+)
+```
+
+### Performance Issues
+
+If experiencing slow performance:
+```python
+# Use selective undetected mode only for protected sites
+if is_protected_site(url):
+    adapter = UndetectedAdapter()
+else:
+    adapter = PlaywrightAdapter()  # Default adapter
+```
+
+## Future Plans
+
+**Note**: In future versions of Crawl4AI, we may enable stealth mode and undetected browser by default to provide better out-of-the-box success rates. For now, users should explicitly enable these features when needed.
+
+## Conclusion
+
+Crawl4AI provides flexible anti-bot solutions:
+
+1. **Start Simple**: Use regular browser + stealth mode for most sites
+2. **Escalate if Needed**: Switch to undetected browser for sophisticated protection
+3. **Combine for Maximum Effect**: Use both features together when facing the toughest challenges
+
+Remember:
+- Always respect robots.txt and website terms of service
+- Use appropriate delays to avoid overwhelming servers
+- Consider the performance trade-offs of each approach
+- Test progressively to find the minimum necessary evasion level
+
+## See Also
+
+- [Advanced Features](advanced-features.md) - Overview of all advanced features
+- [Proxy & Security](proxy-security.md) - Using proxies with anti-bot features
+- [Session Management](session-management.md) - Maintaining sessions across requests
+- [Identity Based Crawling](identity-based-crawling.md) - Additional anti-detection strategies

docs/md_v2/api/arun_many.md

@@ -7,7 +7,7 @@
 ```python
 async def arun_many(
     urls: Union[List[str], List[Any]],
-    config: Optional[CrawlerRunConfig] = None,
+    config: Optional[Union[CrawlerRunConfig, List[CrawlerRunConfig]]] = None,
     dispatcher: Optional[BaseDispatcher] = None,
     ...
 ) -> Union[List[CrawlResult], AsyncGenerator[CrawlResult, None]]:
@@ -15,7 +15,9 @@ async def arun_many(
     Crawl multiple URLs concurrently or in batches.
 
     :param urls: A list of URLs (or tasks) to crawl.
-    :param config: (Optional) A default `CrawlerRunConfig` applying to each crawl.
+    :param config: (Optional) Either:
+        - A single `CrawlerRunConfig` applying to all URLs
+        - A list of `CrawlerRunConfig` objects with url_matcher patterns
     :param dispatcher: (Optional) A concurrency controller (e.g. MemoryAdaptiveDispatcher).
     ...
     :return: Either a list of `CrawlResult` objects, or an async generator if streaming is enabled.
@@ -95,10 +97,70 @@ results = await crawler.arun_many(
 )
 ```
 
+### URL-Specific Configurations
+
+Instead of using one config for all URLs, provide a list of configs with `url_matcher` patterns:
+
+```python
+from crawl4ai import CrawlerRunConfig, MatchMode
+from crawl4ai.processors.pdf import PDFContentScrapingStrategy
+from crawl4ai.extraction_strategy import JsonCssExtractionStrategy
+from crawl4ai.content_filter_strategy import PruningContentFilter
+from crawl4ai.markdown_generation_strategy import DefaultMarkdownGenerator
+
+# PDF files - specialized extraction
+pdf_config = CrawlerRunConfig(
+    url_matcher="*.pdf",
+    scraping_strategy=PDFContentScrapingStrategy()
+)
+
+# Blog/article pages - content filtering
+blog_config = CrawlerRunConfig(
+    url_matcher=["*/blog/*", "*/article/*", "*python.org*"],
+    markdown_generator=DefaultMarkdownGenerator(
+        content_filter=PruningContentFilter(threshold=0.48)
+    )
+)
+
+# Dynamic pages - JavaScript execution
+github_config = CrawlerRunConfig(
+    url_matcher=lambda url: 'github.com' in url,
+    js_code="window.scrollTo(0, 500);"
+)
+
+# API endpoints - JSON extraction
+api_config = CrawlerRunConfig(
+    url_matcher=lambda url: 'api' in url or url.endswith('.json'),
+    # Custome settings for JSON extraction
+)
+
+# Default fallback config
+default_config = CrawlerRunConfig()  # No url_matcher means it never matches except as fallback
+
+# Pass the list of configs - first match wins!
+results = await crawler.arun_many(
+    urls=[
+        "https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf",  # → pdf_config
+        "https://blog.python.org/",  # → blog_config
+        "https://github.com/microsoft/playwright",  # → github_config
+        "https://httpbin.org/json",  # → api_config
+        "https://example.com/"  # → default_config
+    ],
+    config=[pdf_config, blog_config, github_config, api_config, default_config]
+)
+```
+
+**URL Matching Features**:
+- **String patterns**: `"*.pdf"`, `"*/blog/*"`, `"*python.org*"`
+- **Function matchers**: `lambda url: 'api' in url`
+- **Mixed patterns**: Combine strings and functions with `MatchMode.OR` or `MatchMode.AND`
+- **First match wins**: Configs are evaluated in order
+
 **Key Points**:
 - Each URL is processed by the same or separate sessions, depending on the dispatcher’s strategy.
 - `dispatch_result` in each `CrawlResult` (if using concurrency) can hold memory and timing info.  
 - If you need to handle authentication or session IDs, pass them in each individual task or within your run config.
+- **Important**: Always include a default config (without `url_matcher`) as the last item if you want to handle all URLs. Otherwise, unmatched URLs will fail.
 
 ### Return Value
 

docs/md_v2/api/parameters.md

@@ -208,6 +208,71 @@ config = CrawlerRunConfig(
 
 See [Virtual Scroll documentation](../../advanced/virtual-scroll.md) for detailed examples.
 
+---
+
+### I) **URL Matching Configuration**
+
+| **Parameter**          | **Type / Default**           | **What It Does**                                                                                                                    |
+|------------------------|------------------------------|-------------------------------------------------------------------------------------------------------------------------------------|
+| **`url_matcher`**      | `UrlMatcher` (None)          | Pattern(s) to match URLs against. Can be: string (glob), function, or list of mixed types. **None means match ALL URLs**         |
+| **`match_mode`**       | `MatchMode` (MatchMode.OR)   | How to combine multiple matchers in a list: `MatchMode.OR` (any match) or `MatchMode.AND` (all must match)                       |
+
+The `url_matcher` parameter enables URL-specific configurations when used with `arun_many()`:
+
+```python
+from crawl4ai import CrawlerRunConfig, MatchMode
+from crawl4ai.processors.pdf import PDFContentScrapingStrategy
+from crawl4ai.extraction_strategy import JsonCssExtractionStrategy
+
+# Simple string pattern (glob-style)
+pdf_config = CrawlerRunConfig(
+    url_matcher="*.pdf",
+    scraping_strategy=PDFContentScrapingStrategy()
+)
+
+# Multiple patterns with OR logic (default)
+blog_config = CrawlerRunConfig(
+    url_matcher=["*/blog/*", "*/article/*", "*/news/*"],
+    match_mode=MatchMode.OR  # Any pattern matches
+)
+
+# Function matcher
+api_config = CrawlerRunConfig(
+    url_matcher=lambda url: 'api' in url or url.endswith('.json'),
+    # Other settings like extraction_strategy
+)
+
+# Mixed: String + Function with AND logic
+complex_config = CrawlerRunConfig(
+    url_matcher=[
+        lambda url: url.startswith('https://'),  # Must be HTTPS
+        "*.org/*",                               # Must be .org domain
+        lambda url: 'docs' in url                # Must contain 'docs'
+    ],
+    match_mode=MatchMode.AND  # ALL conditions must match
+)
+
+# Combined patterns and functions with AND logic
+secure_docs = CrawlerRunConfig(
+    url_matcher=["https://*", lambda url: '.doc' in url],
+    match_mode=MatchMode.AND  # Must be HTTPS AND contain .doc
+)
+
+# Default config - matches ALL URLs
+default_config = CrawlerRunConfig()  # No url_matcher = matches everything
+```
+
+**UrlMatcher Types:**
+- **None (default)**: When `url_matcher` is None or not set, the config matches ALL URLs
+- **String patterns**: Glob-style patterns like `"*.pdf"`, `"*/api/*"`, `"https://*.example.com/*"`
+- **Functions**: `lambda url: bool` - Custom logic for complex matching
+- **Lists**: Mix strings and functions, combined with `MatchMode.OR` or `MatchMode.AND`
+
+**Important Behavior:**
+- When passing a list of configs to `arun_many()`, URLs are matched against each config's `url_matcher` in order. First match wins!
+- If no config matches a URL and there's no default config (one without `url_matcher`), the URL will fail with "No matching configuration found"
+- Always include a default config as the last item if you want to handle all URLs
+
 ---## 2.2 Helper Methods
 
 Both `BrowserConfig` and `CrawlerRunConfig` provide a `clone()` method to create modified copies:

docs/md_v2/blog/index.md

@@ -20,24 +20,30 @@ Ever wondered why your AI coding assistant struggles with your library despite c
 
 ## Latest Release
 
-### [Crawl4AI v0.7.0 – The Adaptive Intelligence Update](releases/0.7.0.md)
-*January 28, 2025*
+### [Crawl4AI v0.7.3 – The Multi-Config Intelligence Update](releases/0.7.3.md)
+*August 6, 2025*
 
-Crawl4AI v0.7.0 introduces groundbreaking intelligence features that transform how crawlers understand and adapt to websites. This release brings Adaptive Crawling that learns website patterns, Virtual Scroll support for infinite pages, intelligent Link Preview with 3-layer scoring, and the powerful Async URL Seeder for massive URL discovery.
+Crawl4AI v0.7.3 brings smarter URL-specific configurations, flexible Docker deployments, and critical stability improvements. Configure different crawling strategies for different URL patterns in a single batch—perfect for mixed content sites with docs, blogs, and APIs.
 
 Key highlights:
-- **Adaptive Crawling**: Crawlers that learn and adapt to website structures automatically
-- **Virtual Scroll Support**: Complete content extraction from modern infinite scroll pages  
-- **Link Preview**: 3-layer scoring system for intelligent link prioritization
-- **Async URL Seeder**: Discover thousands of URLs in seconds with smart filtering
-- **Performance Boost**: Up to 3x faster with optimized resource handling
+- **Multi-URL Configurations**: Different strategies for different URL patterns in one crawl
+- **Flexible Docker LLM Providers**: Configure providers via environment variables  
+- **Bug Fixes**: Critical stability improvements for production deployments
+- **Documentation Updates**: Clearer examples and improved API documentation
 
-[Read full release notes →](releases/0.7.0.md)
+[Read full release notes →](releases/0.7.3.md)
 
 ---
 
 ## Previous Releases
 
+### [Crawl4AI v0.7.0 – The Adaptive Intelligence Update](releases/0.7.0.md)
+*January 28, 2025*
+
+Introduced groundbreaking intelligence features including Adaptive Crawling, Virtual Scroll support, intelligent Link Preview, and the Async URL Seeder for massive URL discovery.
+
+[Read release notes →](releases/0.7.0.md)
+
 ### [Crawl4AI v0.6.0 – World-Aware Crawling, Pre-Warmed Browsers, and the MCP API](releases/0.6.0.md)
 *December 23, 2024*
 

docs/md_v2/blog/releases/0.7.1.md

@@ -0,0 +1,43 @@
+# 🛠️ Crawl4AI v0.7.1: Minor Cleanup Update
+
+*July 17, 2025 • 2 min read*
+
+---
+
+A small maintenance release that removes unused code and improves documentation.
+
+## 🎯 What's Changed
+
+- **Removed unused StealthConfig** from `crawl4ai/browser_manager.py`
+- **Updated documentation** with better examples and parameter explanations
+- **Fixed virtual scroll configuration** examples in docs
+
+## 🧹 Code Cleanup
+
+Removed unused `StealthConfig` import and configuration that wasn't being used anywhere in the codebase. The project uses its own custom stealth implementation through JavaScript injection instead.
+
+```python
+# Removed unused code:
+from playwright_stealth import StealthConfig
+stealth_config = StealthConfig(...)  # This was never used
+```
+
+## 📖 Documentation Updates
+
+- Fixed adaptive crawling parameter examples
+- Updated session management documentation
+- Corrected virtual scroll configuration examples
+
+## 🚀 Installation
+
+```bash
+pip install crawl4ai==0.7.1
+```
+
+No breaking changes - upgrade directly from v0.7.0.
+
+---
+
+Questions? Issues? 
+- GitHub: [github.com/unclecode/crawl4ai](https://github.com/unclecode/crawl4ai)
+- Discord: [discord.gg/crawl4ai](https://discord.gg/jP8KfhDhyN)

docs/md_v2/blog/releases/0.7.2.md

@@ -0,0 +1,98 @@
+# 🚀 Crawl4AI v0.7.2: CI/CD & Dependency Optimization Update
+
+*July 25, 2025 • 3 min read*
+
+---
+
+This release introduces automated CI/CD pipelines for seamless releases and optimizes dependencies for a lighter, more efficient package.
+
+## 🎯 What's New
+
+### 🔄 Automated Release Pipeline
+- **GitHub Actions CI/CD**: Automated PyPI and Docker Hub releases on tag push
+- **Multi-platform Docker images**: Support for both AMD64 and ARM64 architectures
+- **Version consistency checks**: Ensures tag, package, and Docker versions align
+- **Automated release notes**: GitHub releases created automatically
+
+### 📦 Dependency Optimization
+- **Moved sentence-transformers to optional dependencies**: Significantly reduces default installation size
+- **Lighter Docker images**: Optimized Dockerfile for faster builds and smaller images
+- **Better dependency management**: Core vs. optional dependencies clearly separated
+
+## 🏗️ CI/CD Pipeline
+
+The new automated release process ensures consistent, reliable releases:
+
+```yaml
+# Trigger releases with a simple tag
+git tag v0.7.2
+git push origin v0.7.2
+
+# Automatically:
+# ✅ Validates version consistency
+# ✅ Builds and publishes to PyPI
+# ✅ Builds multi-platform Docker images
+# ✅ Pushes to Docker Hub with proper tags
+# ✅ Creates GitHub release
+```
+
+## 💾 Lighter Installation
+
+Default installation is now significantly smaller:
+
+```bash
+# Core installation (smaller, faster)
+pip install crawl4ai==0.7.2
+
+# With ML features (includes sentence-transformers)
+pip install crawl4ai[transformer]==0.7.2
+
+# Full installation
+pip install crawl4ai[all]==0.7.2
+```
+
+## 🐳 Docker Improvements
+
+Enhanced Docker support with multi-platform images:
+
+```bash
+# Pull the latest version
+docker pull unclecode/crawl4ai:0.7.2
+docker pull unclecode/crawl4ai:latest
+
+# Available tags:
+# - unclecode/crawl4ai:0.7.2 (specific version)
+# - unclecode/crawl4ai:0.7 (minor version)
+# - unclecode/crawl4ai:0 (major version)
+# - unclecode/crawl4ai:latest
+```
+
+## 🔧 Technical Details
+
+### Dependency Changes
+- `sentence-transformers` moved from required to optional dependencies
+- Reduces default installation by ~500MB
+- No impact on functionality when transformer features aren't needed
+
+### CI/CD Configuration
+- GitHub Actions workflows for automated releases
+- Version validation before publishing
+- Parallel PyPI and Docker Hub deployments
+- Automatic tagging strategy for Docker images
+
+## 🚀 Installation
+
+```bash
+pip install crawl4ai==0.7.2
+```
+
+No breaking changes - direct upgrade from v0.7.0 or v0.7.1.
+
+---
+
+Questions? Issues? 
+- GitHub: [github.com/unclecode/crawl4ai](https://github.com/unclecode/crawl4ai)
+- Discord: [discord.gg/crawl4ai](https://discord.gg/jP8KfhDhyN)
+- Twitter: [@unclecode](https://x.com/unclecode)
+
+*P.S. The new CI/CD pipeline will make future releases faster and more reliable. Thanks for your patience as we improve our release process!*

docs/md_v2/blog/releases/0.7.3.md

@@ -0,0 +1,170 @@
+# 🚀 Crawl4AI v0.7.3: The Multi-Config Intelligence Update
+
+*August 6, 2025 • 5 min read*
+
+---
+
+Today I'm releasing Crawl4AI v0.7.3—the Multi-Config Intelligence Update. This release brings smarter URL-specific configurations, flexible Docker deployments, important bug fixes, and documentation improvements that make Crawl4AI more robust and production-ready.
+
+## 🎯 What's New at a Glance
+
+- **Multi-URL Configurations**: Different crawling strategies for different URL patterns in a single batch
+- **Flexible Docker LLM Providers**: Configure LLM providers via environment variables
+- **Bug Fixes**: Resolved several critical issues for better stability
+- **Documentation Updates**: Clearer examples and improved API documentation
+
+## 🎨 Multi-URL Configurations: One Size Doesn't Fit All
+
+**The Problem:** You're crawling a mix of documentation sites, blogs, and API endpoints. Each needs different handling—caching for docs, fresh content for news, structured extraction for APIs. Previously, you'd run separate crawls or write complex conditional logic.
+
+**My Solution:** I implemented URL-specific configurations that let you define different strategies for different URL patterns in a single crawl batch. First match wins, with optional fallback support.
+
+### Technical Implementation
+
+```python
+from crawl4ai import AsyncWebCrawler, CrawlerRunConfig, MatchMode
+
+# Define specialized configs for different content types
+configs = [
+    # Documentation sites - aggressive caching, include links
+    CrawlerRunConfig(
+        url_matcher=["*docs*", "*documentation*"],
+        cache_mode="write",
+        markdown_generator_options={"include_links": True}
+    ),
+    
+    # News/blog sites - fresh content, scroll for lazy loading
+    CrawlerRunConfig(
+        url_matcher=lambda url: 'blog' in url or 'news' in url,
+        cache_mode="bypass",
+        js_code="window.scrollTo(0, document.body.scrollHeight/2);"
+    ),
+    
+    # API endpoints - structured extraction
+    CrawlerRunConfig(
+        url_matcher=["*.json", "*api*"],
+        extraction_strategy=LLMExtractionStrategy(
+            provider="openai/gpt-4o-mini",
+            extraction_type="structured"
+        )
+    ),
+    
+    # Default fallback for everything else
+    CrawlerRunConfig()  # No url_matcher = matches everything
+]
+
+# Crawl multiple URLs with appropriate configs
+async with AsyncWebCrawler() as crawler:
+    results = await crawler.arun_many(
+        urls=[
+            "https://docs.python.org/3/",      # → Uses documentation config
+            "https://blog.python.org/",        # → Uses blog config  
+            "https://api.github.com/users",    # → Uses API config
+            "https://example.com/"             # → Uses default config
+        ],
+        config=configs
+    )
+```
+
+**Matching Capabilities:**
+- **String Patterns**: Wildcards like `"*.pdf"`, `"*/blog/*"`
+- **Function Matchers**: Lambda functions for complex logic
+- **Mixed Matchers**: Combine strings and functions with AND/OR logic
+- **Fallback Support**: Default config when nothing matches
+
+**Expected Real-World Impact:**
+- **Mixed Content Sites**: Handle blogs, docs, and downloads in one crawl
+- **Multi-Domain Crawling**: Different strategies per domain without separate runs
+- **Reduced Complexity**: No more if/else forests in your extraction code
+- **Better Performance**: Each URL gets exactly the processing it needs
+
+## 🐳 Docker: Flexible LLM Provider Configuration
+
+**The Problem:** Hardcoded LLM providers in Docker deployments. Want to switch from OpenAI to Groq? Rebuild and redeploy. Testing different models? Multiple Docker images.
+
+**My Solution:** Configure LLM providers via environment variables. Switch providers without touching code or rebuilding images.
+
+### Deployment Flexibility
+
+```bash
+# Option 1: Direct environment variables
+docker run -d \
+  -e LLM_PROVIDER="groq/llama-3.2-3b-preview" \
+  -e GROQ_API_KEY="your-key" \
+  -p 11235:11235 \
+  unclecode/crawl4ai:latest
+
+# Option 2: Using .llm.env file (recommended for production)
+# Create .llm.env file:
+# LLM_PROVIDER=openai/gpt-4o-mini
+# OPENAI_API_KEY=your-openai-key
+# GROQ_API_KEY=your-groq-key
+
+docker run -d \
+  --env-file .llm.env \
+  -p 11235:11235 \
+  unclecode/crawl4ai:latest
+```
+
+Override per request when needed:
+```python
+# Use default provider from .llm.env
+response = requests.post("http://localhost:11235/crawl", json={
+    "url": "https://example.com",
+    "extraction_strategy": {"type": "llm"}
+})
+
+# Override to use different provider for this specific request
+response = requests.post("http://localhost:11235/crawl", json={
+    "url": "https://complex-page.com",
+    "extraction_strategy": {
+        "type": "llm",
+        "provider": "openai/gpt-4"  # Override default
+    }
+})
+```
+
+**Expected Real-World Impact:**
+- **Cost Optimization**: Use cheaper models for simple tasks, premium for complex
+- **A/B Testing**: Compare provider performance without deployment changes
+- **Fallback Strategies**: Switch providers on-the-fly during outages
+- **Development Flexibility**: Test locally with one provider, deploy with another
+- **Secure Configuration**: Keep API keys in `.llm.env` file, not in commands
+
+## 🔧 Bug Fixes & Improvements
+
+This release includes several important bug fixes that improve stability and reliability:
+
+- **URL Matcher Fallback**: Fixed edge cases in URL pattern matching logic
+- **Memory Management**: Resolved memory leaks in long-running crawl sessions
+- **Sitemap Processing**: Fixed redirect handling in sitemap fetching
+- **Table Extraction**: Improved table detection and extraction accuracy
+- **Error Handling**: Better error messages and recovery from network failures
+
+## 📚 Documentation Enhancements
+
+Based on community feedback, we've updated:
+- Clearer examples for multi-URL configuration
+- Improved CrawlResult documentation with all available fields
+- Fixed typos and inconsistencies across documentation
+- Added real-world URLs in examples for better understanding
+- New comprehensive demo showcasing all v0.7.3 features
+
+## 🙏 Acknowledgments
+
+Thanks to our contributors and the entire community for feedback and bug reports.
+
+## 📚 Resources
+
+- [Full Documentation](https://docs.crawl4ai.com)
+- [GitHub Repository](https://github.com/unclecode/crawl4ai)
+- [Discord Community](https://discord.gg/crawl4ai)
+- [Feature Demo](https://github.com/unclecode/crawl4ai/blob/main/docs/releases_review/demo_v0.7.3.py)
+
+---
+
+*Crawl4AI continues to evolve with your needs. This release makes it smarter, more flexible, and more stable. Try the new multi-config feature and flexible Docker deployment—they're game changers!*
+
+**Happy Crawling! 🕷️**
+
+*- The Crawl4AI Team*

docs/md_v2/core/browser-crawler-config.md

@@ -29,6 +29,7 @@ class BrowserConfig:
         text_mode=False,
         light_mode=False,
         extra_args=None,
+        enable_stealth=False,
         # ... other advanced parameters omitted here
     ):
         ...
@@ -84,6 +85,11 @@ class BrowserConfig:
     - Additional flags for the underlying browser.  
     - E.g. `["--disable-extensions"]`.
 
+11. **`enable_stealth`**:  
+    - If `True`, enables stealth mode using playwright-stealth.  
+    - Modifies browser fingerprints to avoid basic bot detection.  
+    - Default is `False`. Recommended for sites with bot protection.
+
 ### Helper Methods
 
 Both configuration classes provide a `clone()` method to create modified copies:
@@ -209,7 +215,13 @@ class CrawlerRunConfig:
     - The maximum number of concurrent crawl sessions.  
     - Helps prevent overwhelming the system.
 
-14. **`display_mode`**:  
+14. **`url_matcher`** & **`match_mode`**:  
+    - Enable URL-specific configurations when used with `arun_many()`.
+    - Set `url_matcher` to patterns (glob, function, or list) to match specific URLs.
+    - Use `match_mode` (OR/AND) to control how multiple patterns combine.
+    - See [URL-Specific Configurations](../api/arun_many.md#url-specific-configurations) for examples.
+
+15. **`display_mode`**:  
     - The display mode for progress information (`DETAILED`, `BRIEF`, etc.).  
     - Affects how much information is printed during the crawl.
 

docs/md_v2/core/c4a-script.md

@@ -52,11 +52,9 @@ That's it! In just a few lines, you've automated a complete search workflow.
 
 Want to learn by doing? We've got you covered:
 
-**🚀 [Live Demo](https://docs.crawl4ai.com/c4a-script/demo)** - Try C4A-Script in your browser right now!
+**🚀 [Live Demo](https://docs.crawl4ai.com/apps/c4a-script/)** - Try C4A-Script in your browser right now!
 
-**📁 [Tutorial Examples](/examples/c4a_script/)** - Complete examples with source code
-
-**🛠️ [Local Tutorial](/examples/c4a_script/tutorial/)** - Run the interactive tutorial on your machine
+**📁 [Tutorial Examples](https://github.com/unclecode/crawl4ai/blob/main/docs/examples/c4a_script/)** - Complete examples with source code
 
 ### Running the Tutorial Locally
 

docs/md_v2/core/content-selection.md

@@ -350,15 +350,22 @@ if __name__ == "__main__":
 
 ## 6. Scraping Modes
 
-Crawl4AI provides two different scraping strategies for HTML content processing: `WebScrapingStrategy` (BeautifulSoup-based, default) and `LXMLWebScrapingStrategy` (LXML-based). The LXML strategy offers significantly better performance, especially for large HTML documents.
+Crawl4AI uses `LXMLWebScrapingStrategy` (LXML-based) as the default scraping strategy for HTML content processing. This strategy offers excellent performance, especially for large HTML documents.
+
+**Note:** For backward compatibility, `WebScrapingStrategy` is still available as an alias for `LXMLWebScrapingStrategy`.
 
 ```python
 from crawl4ai import AsyncWebCrawler, CrawlerRunConfig, LXMLWebScrapingStrategy
 
 async def main():
-    config = CrawlerRunConfig(
-        scraping_strategy=LXMLWebScrapingStrategy()  # Faster alternative to default BeautifulSoup
+    # Default configuration already uses LXMLWebScrapingStrategy
+    config = CrawlerRunConfig()
+    
+    # Or explicitly specify it if desired
+    config_explicit = CrawlerRunConfig(
+        scraping_strategy=LXMLWebScrapingStrategy()
     )
+    
     async with AsyncWebCrawler() as crawler:
         result = await crawler.arun(
             url="https://example.com", 
@@ -417,21 +424,20 @@ class CustomScrapingStrategy(ContentScrapingStrategy):
 
 ### Performance Considerations
 
-The LXML strategy can be up to 10-20x faster than BeautifulSoup strategy, particularly when processing large HTML documents. However, please note:
+The LXML strategy provides excellent performance, particularly when processing large HTML documents, offering up to 10-20x faster processing compared to BeautifulSoup-based approaches.
 
-1. LXML strategy is currently experimental
-2. In some edge cases, the parsing results might differ slightly from BeautifulSoup
-3. If you encounter any inconsistencies between LXML and BeautifulSoup results, please [raise an issue](https://github.com/codeium/crawl4ai/issues) with a reproducible example
+Benefits of LXML strategy:
+- Fast processing of large HTML documents (especially >100KB)
+- Efficient memory usage
+- Good handling of well-formed HTML
+- Robust table detection and extraction
 
-Choose LXML strategy when:
-- Processing large HTML documents (recommended for >100KB)
-- Performance is critical
-- Working with well-formed HTML
+### Backward Compatibility
 
-Stick to BeautifulSoup strategy (default) when:
-- Maximum compatibility is needed
-- Working with malformed HTML
-- Exact parsing behavior is critical
+For users upgrading from earlier versions:
+- `WebScrapingStrategy` is now an alias for `LXMLWebScrapingStrategy`
+- Existing code using `WebScrapingStrategy` will continue to work without modification
+- No changes are required to your existing code
 
 ---
 

docs/md_v2/core/crawler-result.md

@@ -19,13 +19,15 @@ class MarkdownGenerationResult(BaseModel):
 class CrawlResult(BaseModel):
     url: str
     html: str
+    fit_html: Optional[str] = None
     success: bool
     cleaned_html: Optional[str] = None
     media: Dict[str, List[Dict]] = {}
     links: Dict[str, List[Dict]] = {}
     downloaded_files: Optional[List[str]] = None
+    js_execution_result: Optional[Dict[str, Any]] = None
     screenshot: Optional[str] = None
-    pdf : Optional[bytes] = None
+    pdf: Optional[bytes] = None
     mhtml: Optional[str] = None
     markdown: Optional[Union[str, MarkdownGenerationResult]] = None
     extracted_content: Optional[str] = None
@@ -35,6 +37,12 @@ class CrawlResult(BaseModel):
     response_headers: Optional[dict] = None
     status_code: Optional[int] = None
     ssl_certificate: Optional[SSLCertificate] = None
+    dispatch_result: Optional[DispatchResult] = None
+    redirected_url: Optional[str] = None
+    network_requests: Optional[List[Dict[str, Any]]] = None
+    console_messages: Optional[List[Dict[str, Any]]] = None
+    tables: List[Dict] = Field(default_factory=list)
+
     class Config:
         arbitrary_types_allowed = True
 ```
@@ -45,11 +53,13 @@ class CrawlResult(BaseModel):
 |-------------------------------------------|-----------------------------------------------------------------------------------------------------|
 | **url (`str`)**                           | The final or actual URL crawled (in case of redirects).                                             |
 | **html (`str`)**                          | Original, unmodified page HTML. Good for debugging or custom processing.                            |
+| **fit_html (`Optional[str]`)**            | Preprocessed HTML optimized for extraction and content filtering.                                    |
 | **success (`bool`)**                      | `True` if the crawl completed without major errors, else `False`.                                   |
 | **cleaned_html (`Optional[str]`)**        | Sanitized HTML with scripts/styles removed; can exclude tags if configured via `excluded_tags` etc. |
 | **media (`Dict[str, List[Dict]]`)**       | Extracted media info (images, audio, etc.), each with attributes like `src`, `alt`, `score`, etc.   |
 | **links (`Dict[str, List[Dict]]`)**       | Extracted link data, split by `internal` and `external`. Each link usually has `href`, `text`, etc. |
 | **downloaded_files (`Optional[List[str]]`)** | If `accept_downloads=True` in `BrowserConfig`, this lists the filepaths of saved downloads.         |
+| **js_execution_result (`Optional[Dict[str, Any]]`)** | Results from JavaScript execution during crawling. |
 | **screenshot (`Optional[str]`)**          | Screenshot of the page (base64-encoded) if `screenshot=True`.                                       |
 | **pdf (`Optional[bytes]`)**               | PDF of the page if `pdf=True`.                                                                      |
 | **mhtml (`Optional[str]`)**               | MHTML snapshot of the page if `capture_mhtml=True`. Contains the full page with all resources.      |
@@ -61,6 +71,11 @@ class CrawlResult(BaseModel):
 | **response_headers (`Optional[dict]`)**   | HTTP response headers, if captured.                                                                 |
 | **status_code (`Optional[int]`)**         | HTTP status code (e.g., 200 for OK).                                                                |
 | **ssl_certificate (`Optional[SSLCertificate]`)** | SSL certificate info if `fetch_ssl_certificate=True`.                                               |
+| **dispatch_result (`Optional[DispatchResult]`)** | Additional concurrency and resource usage information when crawling URLs in parallel.               |
+| **redirected_url (`Optional[str]`)**      | The URL after any redirects (different from `url` which is the final URL).                          |
+| **network_requests (`Optional[List[Dict[str, Any]]]`)** | List of network requests, responses, and failures captured during the crawl if `capture_network_requests=True`. |
+| **console_messages (`Optional[List[Dict[str, Any]]]`)** | List of browser console messages captured during the crawl if `capture_console_messages=True`.       |
+| **tables (`List[Dict]`)**                 | Table data extracted from HTML tables with structure `[{headers, rows, caption, summary}]`.           |
 
 ---
 
@@ -172,7 +187,7 @@ Here:
 
 ---
 
-## 5. More Fields: Links, Media, and More
+## 5. More Fields: Links, Media, Tables and More
 
 ### 5.1 `links`
 
@@ -192,7 +207,77 @@ for img in images:
     print("Image URL:", img["src"], "Alt:", img.get("alt"))
 ```
 
-### 5.3 `screenshot`, `pdf`, and `mhtml`
+### 5.3 `tables`
+
+The `tables` field contains structured data extracted from HTML tables found on the crawled page. Tables are analyzed based on various criteria to determine if they are actual data tables (as opposed to layout tables), including:
+
+- Presence of thead and tbody sections
+- Use of th elements for headers
+- Column consistency
+- Text density
+- And other factors
+
+Tables that score above the threshold (default: 7) are extracted and stored in result.tables.
+
+### Accessing Table data:
+```python
+import asyncio
+from crawl4ai import AsyncWebCrawler, CrawlerRunConfig
+
+async def main():
+    async with AsyncWebCrawler() as crawler:
+        result = await crawler.arun(
+            url="https://www.w3schools.com/html/html_tables.asp",
+            config=CrawlerRunConfig(
+                table_score_threshold=7  # Minimum score for table detection
+            )
+        )
+
+        if result.success and result.tables:
+            print(f"Found {len(result.tables)} tables")
+
+            for i, table in enumerate(result.tables):
+                print(f"\nTable {i+1}:")
+                print(f"Caption: {table.get('caption', 'No caption')}")
+                print(f"Headers: {table['headers']}")
+                print(f"Rows: {len(table['rows'])}")
+
+                # Print first few rows as example
+                for j, row in enumerate(table['rows'][:3]):
+                    print(f"  Row {j+1}: {row}")
+
+if __name__ == "__main__":
+    asyncio.run(main())
+
+```
+
+### Configuring Table Extraction:
+
+You can adjust the sensitivity of the table detection algorithm with:
+
+```python
+config = CrawlerRunConfig(
+    table_score_threshold=5  # Lower value = more tables detected (default: 7)
+)
+```
+
+Each extracted table contains: 
+
+- `headers`: Column header names 
+- `rows`: List of rows, each containing cell values
+- `caption`: Table caption text (if available) 
+- `summary`: Table summary attribute (if specified)
+
+### Table Extraction Tips
+
+- Not all HTML tables are extracted - only those detected as "data tables" vs. layout tables.
+- Tables with inconsistent cell counts, nested tables, or those used purely for layout may be skipped.
+- If you're missing tables, try adjusting the `table_score_threshold` to a lower value (default is 7).
+
+The table detection algorithm scores tables based on features like consistent columns, presence of headers, text density, and more. Tables scoring above the threshold are considered data tables worth extracting.
+
+
+### 5.4 `screenshot`, `pdf`, and `mhtml`
 
 If you set `screenshot=True`, `pdf=True`, or `capture_mhtml=True` in **`CrawlerRunConfig`**, then:
 
@@ -213,7 +298,7 @@ if result.mhtml:
 
 The MHTML (MIME HTML) format is particularly useful as it captures the entire web page including all of its resources (CSS, images, scripts, etc.) in a single file, making it perfect for archiving or offline viewing.
 
-### 5.4 `ssl_certificate`
+### 5.5 `ssl_certificate`
 
 If `fetch_ssl_certificate=True`, `result.ssl_certificate` holds details about the site’s SSL cert, such as issuer, validity dates, etc.
 

docs/md_v2/core/docker-deployment.md

@@ -58,15 +58,15 @@ Pull and run images directly from Docker Hub without building locally.
 
 #### 1. Pull the Image
 
-Our latest release candidate is `0.7.0-r1`. Images are built with multi-arch manifests, so Docker automatically pulls the correct version for your system.
+Our latest release is `0.7.3`. Images are built with multi-arch manifests, so Docker automatically pulls the correct version for your system.
 
-> ⚠️ **Important Note**: The `latest` tag currently points to the stable `0.6.0` version. After testing and validation, `0.7.0` (without -r1) will be released and `latest` will be updated. For now, please use `0.7.0-r1` to test the new features.
+> 💡 **Note**: The `latest` tag points to the stable `0.7.3` version.
 
 ```bash
-# Pull the release candidate (for testing new features)
-docker pull unclecode/crawl4ai:0.7.0-r1
+# Pull the latest version
+docker pull unclecode/crawl4ai:0.7.3
 
-# Or pull the current stable version (0.6.0)
+# Or pull using the latest tag
 docker pull unclecode/crawl4ai:latest
 ```
 
@@ -126,7 +126,7 @@ docker stop crawl4ai && docker rm crawl4ai
 #### Docker Hub Versioning Explained
 
 *   **Image Name:** `unclecode/crawl4ai`
-*   **Tag Format:** `LIBRARY_VERSION[-SUFFIX]` (e.g., `0.7.0-r1`)
+*   **Tag Format:** `LIBRARY_VERSION[-SUFFIX]` (e.g., `0.7.3`)
     *   `LIBRARY_VERSION`: The semantic version of the core `crawl4ai` Python library
     *   `SUFFIX`: Optional tag for release candidates (``) and revisions (`r1`)
 *   **`latest` Tag:** Points to the most recent stable version
@@ -154,6 +154,30 @@ cp deploy/docker/.llm.env.example .llm.env
 # Now edit .llm.env and add your API keys
 ```
 
+**Flexible LLM Provider Configuration:**
+
+The Docker setup now supports flexible LLM provider configuration through three methods:
+
+1. **Environment Variable** (Highest Priority): Set `LLM_PROVIDER` to override the default
+   ```bash
+   export LLM_PROVIDER="anthropic/claude-3-opus"
+   # Or in your .llm.env file:
+   # LLM_PROVIDER=anthropic/claude-3-opus
+   ```
+
+2. **API Request Parameter**: Specify provider per request
+   ```json
+   {
+     "url": "https://example.com",
+     "f": "llm",
+     "provider": "groq/mixtral-8x7b"
+   }
+   ```
+
+3. **Config File Default**: Falls back to `config.yml` (default: `openai/gpt-4o-mini`)
+
+The system automatically selects the appropriate API key based on the configured `api_key_env` in the config file.
+
 #### 3. Build and Run with Compose
 
 The `docker-compose.yml` file in the project root provides a simplified approach that automatically handles architecture detection using buildx.
@@ -668,7 +692,7 @@ app:
 
 # Default LLM Configuration
 llm:
-  provider: "openai/gpt-4o-mini"
+  provider: "openai/gpt-4o-mini"  # Can be overridden by LLM_PROVIDER env var
   api_key_env: "OPENAI_API_KEY"
   # api_key: sk-...  # If you pass the API key directly then api_key_env will be ignored
 

docs/md_v2/core/examples.md

@@ -28,11 +28,8 @@ This page provides a comprehensive list of example scripts that demonstrate vari
 | Example | Description | Link |
 |---------|-------------|------|
 | Deep Crawling | An extensive tutorial on deep crawling capabilities, demonstrating BFS and BestFirst strategies, stream vs. non-stream execution, filters, scorers, and advanced configurations. | [View Code](https://github.com/unclecode/crawl4ai/blob/main/docs/examples/deepcrawl_example.py) |
-<<<<<<< HEAD
 | Virtual Scroll | Comprehensive examples for handling virtualized scrolling on sites like Twitter, Instagram. Demonstrates different scrolling scenarios with local test server. | [View Code](https://github.com/unclecode/crawl4ai/blob/main/docs/examples/virtual_scroll_example.py) |
-=======
 | Adaptive Crawling | Demonstrates intelligent crawling that automatically determines when sufficient information has been gathered. | [View Code](https://github.com/unclecode/crawl4ai/blob/main/docs/examples/adaptive_crawling/) |
->>>>>>> feature/progressive-crawling
 | Dispatcher | Shows how to use the crawl dispatcher for advanced workload management. | [View Code](https://github.com/unclecode/crawl4ai/blob/main/docs/examples/dispatcher_example.py) |
 | Storage State | Tutorial on managing browser storage state for persistence. | [View Guide](https://github.com/unclecode/crawl4ai/blob/main/docs/examples/storage_state_tutorial.md) |
 | Network Console Capture | Demonstrates how to capture and analyze network requests and console logs. | [View Code](https://github.com/unclecode/crawl4ai/blob/main/docs/examples/network_console_capture_example.py) |
@@ -57,6 +54,16 @@ This page provides a comprehensive list of example scripts that demonstrate vari
 | Crypto Analysis | Demonstrates how to crawl and analyze cryptocurrency data. | [View Code](https://github.com/unclecode/crawl4ai/blob/main/docs/examples/crypto_analysis_example.py) |
 | SERP API | Demonstrates using Crawl4AI with search engine result pages. | [View Code](https://github.com/unclecode/crawl4ai/blob/main/docs/examples/serp_api_project_11_feb.py) |
 
+## Anti-Bot & Stealth Features
+
+| Example | Description | Link |
+|---------|-------------|------|
+| Stealth Mode Quick Start | Five practical examples showing how to use stealth mode for bypassing basic bot detection. | [View Code](https://github.com/unclecode/crawl4ai/blob/main/docs/examples/stealth_mode_quick_start.py) |
+| Stealth Mode Comprehensive | Comprehensive demonstration of stealth mode features with bot detection testing and comparisons. | [View Code](https://github.com/unclecode/crawl4ai/blob/main/docs/examples/stealth_mode_example.py) |
+| Undetected Browser | Simple example showing how to use the undetected browser adapter. | [View Code](https://github.com/unclecode/crawl4ai/blob/main/docs/examples/hello_world_undetected.py) |
+| Undetected Browser Demo | Basic demo comparing regular and undetected browser modes. | [View Code](https://github.com/unclecode/crawl4ai/blob/main/docs/examples/undetected_simple_demo.py) |
+| Undetected Tests | Advanced tests comparing regular vs undetected browsers on various bot detection services. | [View Folder](https://github.com/unclecode/crawl4ai/tree/main/docs/examples/undetectability/) |
+
 ## Customization & Security
 
 | Example | Description | Link |
@@ -117,4 +124,4 @@ Some examples may require:
 
 ## Contributing New Examples
 
-If you've created an interesting example that demonstrates a unique use case or feature of Crawl4AI, we encourage you to contribute it to our examples collection. Please see our [contribution guidelines](https://github.com/unclecode/crawl4ai/blob/main/CONTRIBUTORS.md) for more information.
+If you've created an interesting example that demonstrates a unique use case or feature of Crawl4AI, we encourage you to contribute it to our examples collection. Please see our [contribution guidelines](https://github.com/unclecode/crawl4ai/blob/main/CONTRIBUTORS.md) for more information.

docs/md_v2/core/installation.md

@@ -18,7 +18,7 @@ crawl4ai-setup
 ```
 
 **What does it do?**
-- Installs or updates required Playwright browsers (Chromium, Firefox, etc.)
+- Installs or updates required browser dependencies for both regular and undetected modes
 - Performs OS-level checks (e.g., missing libs on Linux)
 - Confirms your environment is ready to crawl
 

docs/md_v2/core/link-media.md

@@ -520,7 +520,8 @@ This approach is handy when you still want external links but need to block cert
 
 ### 4.1 Accessing `result.media`
 
-By default, Crawl4AI collects images, audio, video URLs, and data tables it finds on the page. These are stored in `result.media`, a dictionary keyed by media type (e.g., `images`, `videos`, `audio`, `tables`).
+By default, Crawl4AI collects images, audio and video URLs it finds on the page. These are stored in `result.media`, a dictionary keyed by media type (e.g., `images`, `videos`, `audio`).
+**Note: Tables have been moved from `result.media["tables"]` to the new `result.tables` format for better organization and direct access.**
 
 **Basic Example**:
 
@@ -534,14 +535,6 @@ if result.success:
         print(f"           Alt text: {img.get('alt', '')}")
         print(f"           Score: {img.get('score')}")
         print(f"           Description: {img.get('desc', '')}\n")
-    
-    # Get tables
-    tables = result.media.get("tables", [])
-    print(f"Found {len(tables)} data tables in total.")
-    for i, table in enumerate(tables):
-        print(f"[Table {i}] Caption: {table.get('caption', 'No caption')}")
-        print(f"           Columns: {len(table.get('headers', []))}")
-        print(f"           Rows: {len(table.get('rows', []))}")
 ```
 
 **Structure Example**:
@@ -568,19 +561,6 @@ result.media = {
   "audio": [
     # Similar structure but with audio-specific fields
   ],
-  "tables": [
-    {
-      "headers": ["Name", "Age", "Location"],
-      "rows": [
-        ["John Doe", "34", "New York"],
-        ["Jane Smith", "28", "San Francisco"],
-        ["Alex Johnson", "42", "Chicago"]
-      ],
-      "caption": "Employee Directory",
-      "summary": "Directory of company employees"
-    },
-    # More tables if present
-  ]
 }
 ```
 
@@ -608,53 +588,7 @@ crawler_cfg = CrawlerRunConfig(
 
 This setting attempts to discard images from outside the primary domain, keeping only those from the site you’re crawling.
 
-### 3.3 Working with Tables
-
-Crawl4AI can detect and extract structured data from HTML tables. Tables are analyzed based on various criteria to determine if they are actual data tables (as opposed to layout tables), including:
-
-- Presence of thead and tbody sections
-- Use of th elements for headers
-- Column consistency
-- Text density
-- And other factors
-
-Tables that score above the threshold (default: 7) are extracted and stored in `result.media.tables`.
-
-**Accessing Table Data**:
-
-```python
-if result.success:
-    tables = result.media.get("tables", [])
-    print(f"Found {len(tables)} data tables on the page")
-    
-    if tables:
-        # Access the first table
-        first_table = tables[0]
-        print(f"Table caption: {first_table.get('caption', 'No caption')}")
-        print(f"Headers: {first_table.get('headers', [])}")
-        
-        # Print the first 3 rows
-        for i, row in enumerate(first_table.get('rows', [])[:3]):
-            print(f"Row {i+1}: {row}")
-```
-
-**Configuring Table Extraction**:
-
-You can adjust the sensitivity of the table detection algorithm with:
-
-```python
-crawler_cfg = CrawlerRunConfig(
-    table_score_threshold=5  # Lower value = more tables detected (default: 7)
-)
-```
-
-Each extracted table contains:
-- `headers`: Column header names
-- `rows`: List of rows, each containing cell values
-- `caption`: Table caption text (if available)
-- `summary`: Table summary attribute (if specified)
-
-### 3.4 Additional Media Config
+### 4.3 Additional Media Config
 
 - **`screenshot`**: Set to `True` if you want a full-page screenshot stored as `base64` in `result.screenshot`.  
 - **`pdf`**: Set to `True` if you want a PDF version of the page in `result.pdf`.  
@@ -695,7 +629,7 @@ The MHTML format is particularly useful because:
 
 ---
 
-## 4. Putting It All Together: Link & Media Filtering
+## 5. Putting It All Together: Link & Media Filtering
 
 Here’s a combined example demonstrating how to filter out external links, skip certain domains, and exclude external images:
 
@@ -743,7 +677,7 @@ if __name__ == "__main__":
 
 ---
 
-## 5. Common Pitfalls & Tips
+## 6. Common Pitfalls & Tips
 
 1. **Conflicting Flags**:  
    - `exclude_external_links=True` but then also specifying `exclude_social_media_links=True` is typically fine, but understand that the first setting already discards *all* external links. The second becomes somewhat redundant.  
@@ -762,10 +696,3 @@ if __name__ == "__main__":
 ---
 
 **That’s it for Link & Media Analysis!** You’re now equipped to filter out unwanted sites and zero in on the images and videos that matter for your project.
-### Table Extraction Tips
-
-- Not all HTML tables are extracted - only those detected as "data tables" vs. layout tables.
-- Tables with inconsistent cell counts, nested tables, or those used purely for layout may be skipped.
-- If you're missing tables, try adjusting the `table_score_threshold` to a lower value (default is 7).
-
-The table detection algorithm scores tables based on features like consistent columns, presence of headers, text density, and more. Tables scoring above the threshold are considered data tables worth extracting.

docs/md_v2/migration/webscraping-strategy-migration.md

@@ -0,0 +1,92 @@
+# WebScrapingStrategy Migration Guide
+
+## Overview
+
+Crawl4AI has simplified its content scraping architecture. The BeautifulSoup-based `WebScrapingStrategy` has been deprecated in favor of the faster LXML-based implementation. However, **no action is required** - your existing code will continue to work.
+
+## What Changed?
+
+1. **`WebScrapingStrategy` is now an alias** for `LXMLWebScrapingStrategy`
+2. **The BeautifulSoup implementation has been removed** (~1000 lines of redundant code)
+3. **`LXMLWebScrapingStrategy` inherits directly** from `ContentScrapingStrategy`
+4. **Performance remains optimal** with LXML as the sole implementation
+
+## Backward Compatibility
+
+**Your existing code continues to work without any changes:**
+
+```python
+# This still works perfectly
+from crawl4ai import AsyncWebCrawler, CrawlerRunConfig, WebScrapingStrategy
+
+config = CrawlerRunConfig(
+    scraping_strategy=WebScrapingStrategy()  # Works as before
+)
+```
+
+## Migration Options
+
+You have three options:
+
+### Option 1: Do Nothing (Recommended)
+Your code will continue to work. `WebScrapingStrategy` is permanently aliased to `LXMLWebScrapingStrategy`.
+
+### Option 2: Update Imports (Optional)
+For clarity, you can update your imports:
+
+```python
+# Old (still works)
+from crawl4ai import WebScrapingStrategy
+strategy = WebScrapingStrategy()
+
+# New (more explicit)
+from crawl4ai import LXMLWebScrapingStrategy
+strategy = LXMLWebScrapingStrategy()
+```
+
+### Option 3: Use Default Configuration
+Since `LXMLWebScrapingStrategy` is the default, you can omit the strategy parameter:
+
+```python
+# Simplest approach - uses LXMLWebScrapingStrategy by default
+config = CrawlerRunConfig()
+```
+
+## Type Hints
+
+If you use type hints, both work:
+
+```python
+from crawl4ai import WebScrapingStrategy, LXMLWebScrapingStrategy
+
+def process_with_strategy(strategy: WebScrapingStrategy) -> None:
+    # Works with both WebScrapingStrategy and LXMLWebScrapingStrategy
+    pass
+
+# Both are valid
+process_with_strategy(WebScrapingStrategy())
+process_with_strategy(LXMLWebScrapingStrategy())
+```
+
+## Subclassing
+
+If you've subclassed `WebScrapingStrategy`, it continues to work:
+
+```python
+class MyCustomStrategy(WebScrapingStrategy):
+    def __init__(self):
+        super().__init__()
+        # Your custom code
+```
+
+## Performance Benefits
+
+By consolidating to LXML:
+- **10-20x faster** HTML parsing for large documents
+- **Lower memory usage**
+- **Consistent behavior** across all use cases
+- **Simplified maintenance** and bug fixes
+
+## Summary
+
+This change simplifies Crawl4AI's internals while maintaining 100% backward compatibility. Your existing code continues to work, and you get better performance automatically.

mkdocs.yml

@@ -45,6 +45,7 @@ nav:
     - "Lazy Loading": "advanced/lazy-loading.md"
     - "Hooks & Auth": "advanced/hooks-auth.md"
     - "Proxy & Security": "advanced/proxy-security.md"
+    - "Undetected Browser": "advanced/undetected-browser.md"
     - "Session Management": "advanced/session-management.md"
     - "Multi-URL Crawling": "advanced/multi-url-crawling.md"
     - "Crawl Dispatcher": "advanced/crawl-dispatcher.md"

pyproject.toml

@@ -13,38 +13,37 @@ authors = [
     {name = "Unclecode", email = "unclecode@kidocode.com"}
 ]
 dependencies = [
+    "aiofiles>=24.1.0",
+    "aiohttp>=3.11.11",
     "aiosqlite~=0.20",
+    "anyio>=4.0.0",
     "lxml~=5.3",
     "litellm>=1.53.1",
     "numpy>=1.26.0,<3",
     "pillow>=10.4",
     "playwright>=1.49.0",
+    "patchright>=1.49.0",
     "python-dotenv~=1.0",
     "requests~=2.26",
     "beautifulsoup4~=4.12",
     "tf-playwright-stealth>=1.1.0",
     "xxhash~=3.4",
     "rank-bm25~=0.2",
-    "aiofiles>=24.1.0",
     "snowballstemmer~=2.2",
     "pydantic>=2.10",
     "pyOpenSSL>=24.3.0",
     "psutil>=6.1.1",
+    "PyYAML>=6.0",
     "nltk>=3.9.1",
-    "playwright",
     "rich>=13.9.4",
-    "cssselect>=1.2.0",
     "httpx>=0.27.2",
     "httpx[http2]>=0.27.2",
     "fake-useragent>=2.0.3",
     "click>=8.1.7",
-    "pyperclip>=1.8.2",
     "chardet>=5.2.0",
-    "aiohttp>=3.11.11",
     "brotli>=1.1.0",
     "humanize>=4.10.0",
     "lark>=1.2.2",
-    "sentence-transformers>=2.2.0",
     "alphashape>=1.3.1",
     "shapely>=2.0.0"
 ]
@@ -62,8 +61,8 @@ classifiers = [
 [project.optional-dependencies]
 pdf = ["PyPDF2"]  
 torch = ["torch", "nltk", "scikit-learn"]
-transformer = ["transformers", "tokenizers"]
-cosine = ["torch", "transformers", "nltk"]
+transformer = ["transformers", "tokenizers", "sentence-transformers"]
+cosine = ["torch", "transformers", "nltk", "sentence-transformers"]
 sync = ["selenium"]
 all = [
     "PyPDF2",
@@ -72,8 +71,8 @@ all = [
     "scikit-learn",
     "transformers",
     "tokenizers",
-    "selenium",
-    "PyPDF2"  
+    "sentence-transformers",
+    "selenium"
 ]
 
 [project.scripts]

requirements.txt

@@ -1,30 +1,32 @@
 # Note: These requirements are also specified in pyproject.toml
 # This file is kept for development environment setup and compatibility
+aiofiles>=24.1.0
+aiohttp>=3.11.11
 aiosqlite~=0.20
+anyio>=4.0.0
 lxml~=5.3
 litellm>=1.53.1
 numpy>=1.26.0,<3
 pillow>=10.4
 playwright>=1.49.0
+patchright>=1.49.0
 python-dotenv~=1.0
 requests~=2.26
 beautifulsoup4~=4.12
 tf-playwright-stealth>=1.1.0
 xxhash~=3.4
 rank-bm25~=0.2
-aiofiles>=24.1.0
 colorama~=0.4
 snowballstemmer~=2.2
 pydantic>=2.10
 pyOpenSSL>=24.3.0
 psutil>=6.1.1
+PyYAML>=6.0
 nltk>=3.9.1
 rich>=13.9.4
-cssselect>=1.2.0
 chardet>=5.2.0
 brotli>=1.1.0
 httpx[http2]>=0.27.2
-sentence-transformers>=2.2.0
 alphashape>=1.3.1
 shapely>=2.0.0
 

tests/async/test_content_extraction.py

@@ -91,6 +91,17 @@ async def test_css_selector_extraction():
         assert result.markdown
         assert all(heading in result.markdown for heading in ["#", "##", "###"])
 
+@pytest.mark.asyncio
+async def test_base_tag_link_extraction():
+    async with AsyncWebCrawler(verbose=True) as crawler:
+        url = "https://sohamkukreti.github.io/portfolio"
+        result = await crawler.arun(url=url)
+        assert result.success
+        assert result.links
+        assert isinstance(result.links, dict)
+        assert "internal" in result.links
+        assert "external" in result.links
+        assert any("github.com" in x["href"] for x in result.links["external"])
 
 # Entry point for debugging
 if __name__ == "__main__":

tests/async/test_content_scraper_strategy.py

@@ -12,11 +12,8 @@ parent_dir = os.path.dirname(
 sys.path.append(parent_dir)
 __location__ = os.path.realpath(os.path.join(os.getcwd(), os.path.dirname(__file__)))
 
-from crawl4ai.content_scraping_strategy import WebScrapingStrategy
-from crawl4ai.content_scraping_strategy import (
-    WebScrapingStrategy as WebScrapingStrategyCurrent,
-)
-# from crawl4ai.content_scrapping_strategy_current import WebScrapingStrategy as WebScrapingStrategyCurrent
+from crawl4ai.content_scraping_strategy import LXMLWebScrapingStrategy
+# This test compares the same strategy with itself now since WebScrapingStrategy is deprecated
 
 
 @dataclass
@@ -32,8 +29,8 @@ class TestResult:
 
 class StrategyTester:
     def __init__(self):
-        self.new_scraper = WebScrapingStrategy()
-        self.current_scraper = WebScrapingStrategyCurrent()
+        self.new_scraper = LXMLWebScrapingStrategy()
+        self.current_scraper = LXMLWebScrapingStrategy()  # Same strategy now
         with open(__location__ + "/sample_wikipedia.html", "r", encoding="utf-8") as f:
             self.WIKI_HTML = f.read()
         self.results = {"new": [], "current": []}

tests/browser/test_profiles.py

@@ -10,11 +10,13 @@ import sys
 import uuid
 import shutil
 
+from crawl4ai import BrowserProfiler
+from crawl4ai.browser_manager import BrowserManager
+
 # Add the project root to Python path if running directly
 if __name__ == "__main__":
     sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), '../..')))
 
-from crawl4ai.browser import BrowserManager, BrowserProfileManager
 from crawl4ai.async_configs import BrowserConfig, CrawlerRunConfig
 from crawl4ai.async_logger import AsyncLogger
 
@@ -25,7 +27,7 @@ async def test_profile_creation():
     """Test creating and managing browser profiles."""
     logger.info("Testing profile creation and management", tag="TEST")
     
-    profile_manager = BrowserProfileManager(logger=logger)
+    profile_manager = BrowserProfiler(logger=logger)
     
     try:
         # List existing profiles
@@ -83,7 +85,7 @@ async def test_profile_with_browser():
     """Test using a profile with a browser."""
     logger.info("Testing using a profile with a browser", tag="TEST")
     
-    profile_manager = BrowserProfileManager(logger=logger)
+    profile_manager = BrowserProfiler(logger=logger)
     test_profile_name = f"test-browser-profile-{uuid.uuid4().hex[:8]}"
     profile_path = None
     
@@ -101,6 +103,8 @@ async def test_profile_with_browser():
         # Now use this profile with a browser
         browser_config = BrowserConfig(
             user_data_dir=profile_path,
+            use_managed_browser=True,
+            use_persistent_context=True,
             headless=True
         )
         

tests/check_dependencies.py

@@ -0,0 +1,344 @@
+#!/usr/bin/env python3
+"""
+Dependency checker for Crawl4AI
+Analyzes imports in the codebase and shows which files use them
+"""
+
+import ast
+import os
+import sys
+from pathlib import Path
+from typing import Set, Dict, List, Tuple
+from collections import defaultdict
+import re
+import toml
+
+# Standard library modules to ignore
+STDLIB_MODULES = {
+    'abc', 'argparse', 'asyncio', 'base64', 'collections', 'concurrent', 'contextlib',
+    'copy', 'datetime', 'decimal', 'email', 'enum', 'functools', 'glob', 'hashlib',
+    'http', 'importlib', 'io', 'itertools', 'json', 'logging', 'math', 'mimetypes',
+    'multiprocessing', 'os', 'pathlib', 'pickle', 'platform', 'pprint', 'random',
+    're', 'shutil', 'signal', 'socket', 'sqlite3', 'string', 'subprocess', 'sys',
+    'tempfile', 'threading', 'time', 'traceback', 'typing', 'unittest', 'urllib',
+    'uuid', 'warnings', 'weakref', 'xml', 'zipfile', 'dataclasses', 'secrets',
+    'statistics', 'textwrap', 'queue', 'csv', 'gzip', 'tarfile', 'configparser',
+    'inspect', 'operator', 'struct', 'binascii', 'codecs', 'locale', 'gc',
+    'atexit', 'builtins', 'html', 'errno', 'fcntl', 'pwd', 'grp', 'resource',
+    'termios', 'tty', 'pty', 'select', 'selectors', 'ssl', 'zlib', 'bz2',
+    'lzma', 'types', 'copy', 'pydoc', 'profile', 'cProfile', 'timeit',
+    'trace', 'doctest', 'pdb', 'contextvars', 'dataclasses', 'graphlib',
+    'zoneinfo', 'tomllib', 'cgi', 'wsgiref', 'fileinput', 'linecache',
+    'tokenize', 'tabnanny', 'compileall', 'dis', 'pickletools', 'formatter',
+    '__future__', 'array', 'ctypes', 'heapq', 'bisect', 'array', 'weakref',
+    'types', 'copy', 'pprint', 'repr', 'numbers', 'cmath', 'fractions',
+    'statistics', 'itertools', 'functools', 'operator', 'pathlib', 'fileinput',
+    'stat', 'filecmp', 'tempfile', 'glob', 'fnmatch', 'linecache', 'shutil',
+    'pickle', 'copyreg', 'shelve', 'marshal', 'dbm', 'sqlite3', 'zlib', 'gzip',
+    'bz2', 'lzma', 'zipfile', 'tarfile', 'configparser', 'netrc', 'xdrlib',
+    'plistlib', 'hashlib', 'hmac', 'secrets', 'os', 'io', 'time', 'argparse',
+    'getopt', 'logging', 'getpass', 'curses', 'platform', 'errno', 'ctypes',
+    'threading', 'multiprocessing', 'concurrent', 'subprocess', 'sched', 'queue',
+    'contextvars', 'asyncio', 'socket', 'ssl', 'email', 'json', 'mailcap',
+    'mailbox', 'mimetypes', 'base64', 'binhex', 'binascii', 'quopri', 'uu',
+    'html', 'xml', 'webbrowser', 'cgi', 'cgitb', 'wsgiref', 'urllib', 'http',
+    'ftplib', 'poplib', 'imaplib', 'nntplib', 'smtplib', 'smtpd', 'telnetlib',
+    'uuid', 'socketserver', 'xmlrpc', 'ipaddress', 'audioop', 'aifc', 'sunau',
+    'wave', 'chunk', 'colorsys', 'imghdr', 'sndhdr', 'ossaudiodev', 'gettext',
+    'locale', 'turtle', 'cmd', 'shlex', 'tkinter', 'typing', 'pydoc', 'doctest',
+    'unittest', 'test', '2to3', 'distutils', 'venv', 'ensurepip', 'zipapp',
+    'py_compile', 'compileall', 'dis', 'pickletools', 'pdb', 'timeit', 'trace',
+    'tracemalloc', 'warnings', 'faulthandler', 'pdb', 'dataclasses', 'cgi', 
+    'cgitb', 'chunk', 'crypt', 'imghdr', 'mailcap', 'nis', 'nntplib', 'optparse',
+    'ossaudiodev', 'pipes', 'smtpd', 'sndhdr', 'spwd', 'sunau', 'telnetlib',
+    'uu', 'xdrlib', 'msilib', 'pstats', 'rlcompleter', 'tkinter', 'ast'
+}
+
+# Known package name mappings (import name -> package name)
+PACKAGE_MAPPINGS = {
+    'bs4': 'beautifulsoup4',
+    'PIL': 'pillow',
+    'cv2': 'opencv-python',
+    'sklearn': 'scikit-learn',
+    'yaml': 'PyYAML',
+    'OpenSSL': 'pyOpenSSL',
+    'sqlalchemy': 'SQLAlchemy',
+    'playwright': 'playwright',
+    'patchright': 'patchright',
+    'dotenv': 'python-dotenv',
+    'fake_useragent': 'fake-useragent',
+    'playwright_stealth': 'tf-playwright-stealth',
+    'sentence_transformers': 'sentence-transformers',
+    'rank_bm25': 'rank-bm25',
+    'snowballstemmer': 'snowballstemmer',
+    'PyPDF2': 'PyPDF2',
+    'pdf2image': 'pdf2image',
+}
+
+
+class ImportVisitor(ast.NodeVisitor):
+    """AST visitor to extract imports from Python files"""
+    
+    def __init__(self):
+        self.imports = {}  # Changed to dict to store line numbers
+        self.from_imports = {}
+    
+    def visit_Import(self, node):
+        for alias in node.names:
+            module_name = alias.name.split('.')[0]
+            if module_name not in self.imports:
+                self.imports[module_name] = []
+            self.imports[module_name].append(node.lineno)
+    
+    def visit_ImportFrom(self, node):
+        if node.module and node.level == 0:  # absolute imports only
+            module_name = node.module.split('.')[0]
+            if module_name not in self.from_imports:
+                self.from_imports[module_name] = []
+            self.from_imports[module_name].append(node.lineno)
+
+
+def extract_imports_from_file(filepath: Path) -> Dict[str, List[int]]:
+    """Extract all imports from a Python file with line numbers"""
+    all_imports = {}
+    
+    try:
+        with open(filepath, 'r', encoding='utf-8') as f:
+            content = f.read()
+        
+        tree = ast.parse(content)
+        visitor = ImportVisitor()
+        visitor.visit(tree)
+        
+        # Merge imports and from_imports
+        for module, lines in visitor.imports.items():
+            if module not in all_imports:
+                all_imports[module] = []
+            all_imports[module].extend(lines)
+            
+        for module, lines in visitor.from_imports.items():
+            if module not in all_imports:
+                all_imports[module] = []
+            all_imports[module].extend(lines)
+        
+    except Exception as e:
+        # Silently skip files that can't be parsed
+        pass
+    
+    return all_imports
+
+
+def get_codebase_imports_with_files(root_dir: Path) -> Dict[str, List[Tuple[str, List[int]]]]:
+    """Get all imports from the crawl4ai library and docs folders with file locations and line numbers"""
+    import_to_files = defaultdict(list)
+    
+    # Only scan crawl4ai library folder and docs folder
+    target_dirs = [
+        root_dir / 'crawl4ai',
+        root_dir / 'docs'
+    ]
+    
+    for target_dir in target_dirs:
+        if not target_dir.exists():
+            continue
+            
+        for py_file in target_dir.rglob('*.py'):
+            # Skip __pycache__ directories
+            if '__pycache__' in py_file.parts:
+                continue
+            
+            # Skip setup.py and similar files
+            if py_file.name in ['setup.py', 'setup.cfg', 'conf.py']:
+                continue
+                
+            imports = extract_imports_from_file(py_file)
+            
+            # Map each import to the file and line numbers
+            for imp, line_numbers in imports.items():
+                relative_path = py_file.relative_to(root_dir)
+                import_to_files[imp].append((str(relative_path), sorted(line_numbers)))
+    
+    return dict(import_to_files)
+
+
+def get_declared_dependencies() -> Set[str]:
+    """Get declared dependencies from pyproject.toml and requirements.txt"""
+    declared = set()
+    
+    # Read from pyproject.toml
+    if Path('pyproject.toml').exists():
+        with open('pyproject.toml', 'r') as f:
+            data = toml.load(f)
+        
+        # Get main dependencies
+        deps = data.get('project', {}).get('dependencies', [])
+        for dep in deps:
+            # Parse dependency string (e.g., "numpy>=1.26.0,<3")
+            match = re.match(r'^([a-zA-Z0-9_-]+)', dep)
+            if match:
+                pkg_name = match.group(1).lower()
+                declared.add(pkg_name)
+        
+        # Get optional dependencies
+        optional = data.get('project', {}).get('optional-dependencies', {})
+        for group, deps in optional.items():
+            for dep in deps:
+                match = re.match(r'^([a-zA-Z0-9_-]+)', dep)
+                if match:
+                    pkg_name = match.group(1).lower()
+                    declared.add(pkg_name)
+    
+    # Also check requirements.txt as backup
+    if Path('requirements.txt').exists():
+        with open('requirements.txt', 'r') as f:
+            for line in f:
+                line = line.strip()
+                if line and not line.startswith('#'):
+                    match = re.match(r'^([a-zA-Z0-9_-]+)', line)
+                    if match:
+                        pkg_name = match.group(1).lower()
+                        declared.add(pkg_name)
+    
+    return declared
+
+
+def normalize_package_name(name: str) -> str:
+    """Normalize package name for comparison"""
+    # Handle known mappings first
+    if name in PACKAGE_MAPPINGS:
+        return PACKAGE_MAPPINGS[name].lower()
+    
+    # Basic normalization
+    return name.lower().replace('_', '-')
+
+
+def check_missing_dependencies():
+    """Main function to check for missing dependencies"""
+    print("🔍 Analyzing crawl4ai library and docs folders...\n")
+    
+    # Get all imports with their file locations
+    root_dir = Path('.')
+    import_to_files = get_codebase_imports_with_files(root_dir)
+    
+    # Get declared dependencies
+    declared_deps = get_declared_dependencies()
+    
+    # Normalize declared dependencies
+    normalized_declared = {normalize_package_name(dep) for dep in declared_deps}
+    
+    # Categorize imports
+    external_imports = {}
+    local_imports = {}
+    
+    # Known local packages
+    local_packages = {'crawl4ai'}
+    
+    for imp, file_info in import_to_files.items():
+        # Skip standard library
+        if imp in STDLIB_MODULES:
+            continue
+            
+        # Check if it's a local import
+        if any(imp.startswith(local) for local in local_packages):
+            local_imports[imp] = file_info
+        else:
+            external_imports[imp] = file_info
+    
+    # Check which external imports are not declared
+    not_declared = {}
+    declared_imports = {}
+    
+    for imp, file_info in external_imports.items():
+        normalized_imp = normalize_package_name(imp)
+        
+        # Check if import is covered by declared dependencies
+        found = False
+        for declared in normalized_declared:
+            if normalized_imp == declared or normalized_imp.startswith(declared + '.') or declared.startswith(normalized_imp):
+                found = True
+                break
+        
+        if found:
+            declared_imports[imp] = file_info
+        else:
+            not_declared[imp] = file_info
+    
+    # Print results
+    print(f"📊 Summary:")
+    print(f"  - Total unique imports: {len(import_to_files)}")
+    print(f"  - External imports: {len(external_imports)}")
+    print(f"  - Declared dependencies: {len(declared_deps)}")
+    print(f"  - External imports NOT in dependencies: {len(not_declared)}\n")
+    
+    if not_declared:
+        print("❌ External imports NOT declared in pyproject.toml or requirements.txt:\n")
+        
+        # Sort by import name
+        for imp in sorted(not_declared.keys()):
+            file_info = not_declared[imp]
+            print(f"  📦 {imp}")
+            if imp in PACKAGE_MAPPINGS:
+                print(f"     → Package name: {PACKAGE_MAPPINGS[imp]}")
+            
+            # Show up to 3 files that use this import
+            for i, (file_path, line_numbers) in enumerate(file_info[:3]):
+                # Format line numbers for clickable output
+                if len(line_numbers) == 1:
+                    print(f"     - {file_path}:{line_numbers[0]}")
+                else:
+                    # Show first few line numbers
+                    line_str = ','.join(str(ln) for ln in line_numbers[:3])
+                    if len(line_numbers) > 3:
+                        line_str += f"... ({len(line_numbers)} imports)"
+                    print(f"     - {file_path}: lines {line_str}")
+            
+            if len(file_info) > 3:
+                print(f"     ... and {len(file_info) - 3} more files")
+            print()
+    
+    # Check for potentially unused dependencies
+    print("\n🔎 Checking declared dependencies usage...\n")
+    
+    # Get all used external packages
+    used_packages = set()
+    for imp in external_imports.keys():
+        normalized = normalize_package_name(imp)
+        used_packages.add(normalized)
+    
+    # Find unused
+    unused = []
+    for dep in declared_deps:
+        normalized_dep = normalize_package_name(dep)
+        
+        # Check if any import uses this dependency
+        found_usage = False
+        for used in used_packages:
+            if used == normalized_dep or used.startswith(normalized_dep) or normalized_dep.startswith(used):
+                found_usage = True
+                break
+        
+        if not found_usage:
+            # Some packages are commonly unused directly
+            indirect_deps = {'wheel', 'setuptools', 'pip', 'colorama', 'certifi', 'packaging', 'urllib3'}
+            if normalized_dep not in indirect_deps:
+                unused.append(dep)
+    
+    if unused:
+        print("⚠️  Declared dependencies with NO imports found:")
+        for dep in sorted(unused):
+            print(f"  - {dep}")
+        print("\n  Note: These might be used indirectly or by other dependencies")
+    else:
+        print("✅ All declared dependencies have corresponding imports")
+    
+    print("\n" + "="*60)
+    print("💡 How to use this report:")
+    print("  1. Check each ❌ import to see if it's legitimate")
+    print("  2. If legitimate, add the package to pyproject.toml")
+    print("  3. If it's an internal module or typo, fix the import")
+    print("  4. Review unused dependencies - remove if truly not needed")
+    print("="*60)
+
+
+if __name__ == '__main__':
+    check_missing_dependencies()

tests/docker/simple_api_test.py

@@ -168,7 +168,7 @@ class SimpleApiTester:
         print("\n=== CORE APIs ===")
         
         test_url = "https://example.com"
-        
+        test_raw_html_url = "raw://<html><body><h1>Hello, World!</h1></body></html>"
         # Test markdown endpoint
         md_payload = {
             "url": test_url,
@@ -180,6 +180,17 @@ class SimpleApiTester:
         # print(result['data'].get('markdown', ''))
         self.print_result(result)
         
+        # Test markdown endpoint with raw HTML
+        raw_md_payload = {
+            "url": test_raw_html_url,
+            "f": "fit",
+            "q": "test query",
+            "c": "0"
+        }
+        result = self.test_post_endpoint("/md", raw_md_payload)
+        self.print_result(result)
+        
+
         # Test HTML endpoint
         html_payload = {"url": test_url}
         result = self.test_post_endpoint("/html", html_payload)
@@ -215,6 +226,15 @@ class SimpleApiTester:
         result = self.test_post_endpoint("/crawl", crawl_payload)
         self.print_result(result)
         
+        # Test crawl endpoint with raw HTML
+        crawl_payload = {
+            "urls": [test_raw_html_url],
+            "browser_config": {},
+            "crawler_config": {}
+        }
+        result = self.test_post_endpoint("/crawl", crawl_payload)
+        self.print_result(result)
+
         # Test config dump
         config_payload = {"code": "CrawlerRunConfig()"}
         result = self.test_post_endpoint("/config/dump", config_payload)

tests/docker/test_docker.py

@@ -74,7 +74,7 @@ async def test_direct_api():
     # Make direct API call
     async with httpx.AsyncClient() as client:
         response = await client.post(
-            "http://localhost:8000/crawl",
+            "http://localhost:11235/crawl",
             json=request_data,
             timeout=300
         )
@@ -100,13 +100,24 @@ async def test_direct_api():
 
     async with httpx.AsyncClient() as client:
         response = await client.post(
-            "http://localhost:8000/crawl",
+            "http://localhost:11235/crawl",
             json=request_data
         )
         assert response.status_code == 200
         result = response.json()
         print("Structured extraction result:", result["success"])
 
+    # Test 3: Raw HTML
+    request_data["urls"] = ["raw://<html><body><h1>Hello, World!</h1><a href='https://example.com'>Example</a></body></html>"]
+    async with httpx.AsyncClient() as client:
+        response = await client.post(
+            "http://localhost:11235/crawl",
+            json=request_data
+        )
+        assert response.status_code == 200
+        result = response.json()
+        print("Raw HTML result:", result["success"])
+
     # Test 3: Get schema
     # async with httpx.AsyncClient() as client:
     #     response = await client.get("http://localhost:8000/schema")
@@ -118,7 +129,7 @@ async def test_with_client():
     """Test using the Crawl4AI Docker client SDK"""
     print("\n=== Testing Client SDK ===")
     
-    async with Crawl4aiDockerClient(verbose=True) as client:
+    async with Crawl4aiDockerClient(base_url="http://localhost:11235", verbose=True) as client:
         # Test 1: Basic crawl
         browser_config = BrowserConfig(headless=True)
         crawler_config = CrawlerRunConfig(

tests/docker_example.py

@@ -6,28 +6,22 @@ import base64
 import os
 from typing import Dict, Any
 
-
 class Crawl4AiTester:
-    def __init__(self, base_url: str = "http://localhost:11235", api_token: str = None):
+    def __init__(self, base_url: str = "http://localhost:11235"):
         self.base_url = base_url
-        self.api_token = api_token or os.getenv(
-            "CRAWL4AI_API_TOKEN"
-        )  # Check environment variable as fallback
-        self.headers = (
-            {"Authorization": f"Bearer {self.api_token}"} if self.api_token else {}
-        )
+
 
     def submit_and_wait(
         self, request_data: Dict[str, Any], timeout: int = 300
     ) -> Dict[str, Any]:
-        # Submit crawl job
+        # Submit crawl job using async endpoint
         response = requests.post(
-            f"{self.base_url}/crawl", json=request_data, headers=self.headers
+            f"{self.base_url}/crawl/job", json=request_data
         )
-        if response.status_code == 403:
-            raise Exception("API token is invalid or missing")
-        task_id = response.json()["task_id"]
-        print(f"Task ID: {task_id}")
+        response.raise_for_status()
+        job_response = response.json()
+        task_id = job_response["task_id"]
+        print(f"Submitted job with task_id: {task_id}")
 
         # Poll for result
         start_time = time.time()
@@ -38,8 +32,9 @@ class Crawl4AiTester:
                 )
 
             result = requests.get(
-                f"{self.base_url}/task/{task_id}", headers=self.headers
+                f"{self.base_url}/crawl/job/{task_id}"
             )
+            result.raise_for_status()
             status = result.json()
 
             if status["status"] == "failed":
@@ -52,10 +47,10 @@ class Crawl4AiTester:
             time.sleep(2)
 
     def submit_sync(self, request_data: Dict[str, Any]) -> Dict[str, Any]:
+        # Use synchronous crawl endpoint
         response = requests.post(
-            f"{self.base_url}/crawl_sync",
+            f"{self.base_url}/crawl",
             json=request_data,
-            headers=self.headers,
             timeout=60,
         )
         if response.status_code == 408:
@@ -66,9 +61,8 @@ class Crawl4AiTester:
 
 def test_docker_deployment(version="basic"):
     tester = Crawl4AiTester(
-        # base_url="http://localhost:11235" ,
-        base_url="https://crawl4ai-sby74.ondigitalocean.app",
-        api_token="test",
+        base_url="http://localhost:11235",
+        #base_url="https://crawl4ai-sby74.ondigitalocean.app",
     )
     print(f"Testing Crawl4AI Docker {version} version")
 
@@ -88,63 +82,60 @@ def test_docker_deployment(version="basic"):
 
     # Test cases based on version
     test_basic_crawl(tester)
-    test_basic_crawl(tester)
     test_basic_crawl_sync(tester)
 
-    # if version in ["full", "transformer"]:
-    #     test_cosine_extraction(tester)
+    if version in ["full", "transformer"]:
+        test_cosine_extraction(tester)
 
-    # test_js_execution(tester)
-    # test_css_selector(tester)
-    # test_structured_extraction(tester)
-    # test_llm_extraction(tester)
-    # test_llm_with_ollama(tester)
-    # test_screenshot(tester)
+    test_js_execution(tester)
+    test_css_selector(tester)
+    test_structured_extraction(tester)
+    test_llm_extraction(tester)
+    test_llm_with_ollama(tester)
+    test_screenshot(tester)
 
 
 def test_basic_crawl(tester: Crawl4AiTester):
-    print("\n=== Testing Basic Crawl ===")
+    print("\n=== Testing Basic Crawl (Async) ===")
     request = {
         "urls": ["https://www.nbcnews.com/business"],
-        "priority": 10,
-        "session_id": "test",
     }
 
     result = tester.submit_and_wait(request)
-    print(f"Basic crawl result length: {len(result['result']['markdown'])}")
+    print(f"Basic crawl result count: {len(result['result']['results'])}")
     assert result["result"]["success"]
-    assert len(result["result"]["markdown"]) > 0
+    assert len(result["result"]["results"]) > 0
+    assert len(result["result"]["results"][0]["markdown"]) > 0
 
 
 def test_basic_crawl_sync(tester: Crawl4AiTester):
     print("\n=== Testing Basic Crawl (Sync) ===")
     request = {
         "urls": ["https://www.nbcnews.com/business"],
-        "priority": 10,
-        "session_id": "test",
     }
 
     result = tester.submit_sync(request)
-    print(f"Basic crawl result length: {len(result['result']['markdown'])}")
-    assert result["status"] == "completed"
-    assert result["result"]["success"]
-    assert len(result["result"]["markdown"]) > 0
+    print(f"Basic crawl result count: {len(result['results'])}")
+    assert result["success"]
+    assert len(result["results"]) > 0
+    assert len(result["results"][0]["markdown"]) > 0
 
 
 def test_js_execution(tester: Crawl4AiTester):
     print("\n=== Testing JS Execution ===")
     request = {
         "urls": ["https://www.nbcnews.com/business"],
-        "priority": 8,
-        "js_code": [
-            "const loadMoreButton = Array.from(document.querySelectorAll('button')).find(button => button.textContent.includes('Load More')); loadMoreButton && loadMoreButton.click();"
-        ],
-        "wait_for": "article.tease-card:nth-child(10)",
-        "crawler_params": {"headless": True},
+        "browser_config": {"headless": True},
+        "crawler_config": {
+            "js_code": [
+                "const loadMoreButton = Array.from(document.querySelectorAll('button')).find(button => button.textContent.includes('Load More')); if(loadMoreButton) loadMoreButton.click();"
+            ],
+            "wait_for": "wide-tease-item__wrapper df flex-column flex-row-m flex-nowrap-m enable-new-sports-feed-mobile-design(10)"
+        }
     }
 
     result = tester.submit_and_wait(request)
-    print(f"JS execution result length: {len(result['result']['markdown'])}")
+    print(f"JS execution result count: {len(result['result']['results'])}")
     assert result["result"]["success"]
 
 
@@ -152,51 +143,78 @@ def test_css_selector(tester: Crawl4AiTester):
     print("\n=== Testing CSS Selector ===")
     request = {
         "urls": ["https://www.nbcnews.com/business"],
-        "priority": 7,
-        "css_selector": ".wide-tease-item__description",
-        "crawler_params": {"headless": True},
-        "extra": {"word_count_threshold": 10},
+        "browser_config": {"headless": True},
+        "crawler_config": {
+            "css_selector": ".wide-tease-item__description",
+            "word_count_threshold": 10
+        }
     }
 
     result = tester.submit_and_wait(request)
-    print(f"CSS selector result length: {len(result['result']['markdown'])}")
+    print(f"CSS selector result count: {len(result['result']['results'])}")
     assert result["result"]["success"]
 
 
 def test_structured_extraction(tester: Crawl4AiTester):
     print("\n=== Testing Structured Extraction ===")
     schema = {
-        "name": "Coinbase Crypto Prices",
-        "baseSelector": ".cds-tableRow-t45thuk",
-        "fields": [
-            {
-                "name": "crypto",
-                "selector": "td:nth-child(1) h2",
-                "type": "text",
-            },
-            {
-                "name": "symbol",
-                "selector": "td:nth-child(1) p",
-                "type": "text",
-            },
-            {
-                "name": "price",
-                "selector": "td:nth-child(2)",
-                "type": "text",
-            },
-        ],
+  "name": "Cryptocurrency Prices",
+  "baseSelector": "table[data-testid=\"prices-table\"] tbody tr",
+  "fields": [
+    {
+      "name": "asset_name",
+      "selector": "td:nth-child(2) p.cds-headline-h4steop",
+      "type": "text"
+    },
+    {
+      "name": "asset_symbol",
+      "selector": "td:nth-child(2) p.cds-label2-l1sm09ec",
+      "type": "text"
+    },
+    {
+      "name": "asset_image_url",
+      "selector": "td:nth-child(2) img[alt=\"Asset Symbol\"]",
+      "type": "attribute",
+      "attribute": "src"
+    },
+    {
+      "name": "asset_url",
+      "selector": "td:nth-child(2) a[aria-label^=\"Asset page for\"]",
+      "type": "attribute",
+      "attribute": "href"
+    },
+    {
+      "name": "price",
+      "selector": "td:nth-child(3) div.cds-typographyResets-t6muwls.cds-body-bwup3gq",
+      "type": "text"
+    },
+    {
+      "name": "change",
+      "selector": "td:nth-child(7) p.cds-body-bwup3gq",
+      "type": "text"
     }
+  ]
+}
+
 
     request = {
         "urls": ["https://www.coinbase.com/explore"],
-        "priority": 9,
-        "extraction_config": {"type": "json_css", "params": {"schema": schema}},
+        "crawler_config": {
+            "type": "CrawlerRunConfig",
+            "params": {
+                "extraction_strategy": {
+                    "type": "JsonCssExtractionStrategy",
+                    "params": {"schema": schema}
+                }
+            }
+        }
     }
 
     result = tester.submit_and_wait(request)
-    extracted = json.loads(result["result"]["extracted_content"])
+    extracted = json.loads(result["result"]["results"][0]["extracted_content"])
     print(f"Extracted {len(extracted)} items")
-    print("Sample item:", json.dumps(extracted[0], indent=2))
+    if extracted:
+        print("Sample item:", json.dumps(extracted[0], indent=2))
     assert result["result"]["success"]
     assert len(extracted) > 0
 
@@ -206,43 +224,54 @@ def test_llm_extraction(tester: Crawl4AiTester):
     schema = {
         "type": "object",
         "properties": {
-            "model_name": {
+            "asset_name": {
                 "type": "string",
-                "description": "Name of the OpenAI model.",
+                "description": "Name of the asset.",
             },
-            "input_fee": {
+            "price": {
                 "type": "string",
-                "description": "Fee for input token for the OpenAI model.",
+                "description": "Price of the asset.",
             },
-            "output_fee": {
+            "change": {
                 "type": "string",
-                "description": "Fee for output token for the OpenAI model.",
+                "description": "Change in price of the asset.",
             },
         },
-        "required": ["model_name", "input_fee", "output_fee"],
+        "required": ["asset_name", "price", "change"],
     }
 
     request = {
-        "urls": ["https://openai.com/api/pricing"],
-        "priority": 8,
-        "extraction_config": {
-            "type": "llm",
+        "urls": ["https://www.coinbase.com/en-in/explore"],
+        "browser_config": {},
+        "crawler_config": {
+            "type": "CrawlerRunConfig",
             "params": {
-                "provider": "openai/gpt-4o-mini",
-                "api_token": os.getenv("OPENAI_API_KEY"),
-                "schema": schema,
-                "extraction_type": "schema",
-                "instruction": """From the crawled content, extract all mentioned model names along with their fees for input and output tokens.""",
-            },
-        },
-        "crawler_params": {"word_count_threshold": 1},
+                "extraction_strategy": {
+                    "type": "LLMExtractionStrategy",
+                    "params": {
+                        "llm_config": {
+                            "type": "LLMConfig",
+                            "params": {
+                                "provider": "gemini/gemini-2.5-flash",
+                                "api_token": os.getenv("GEMINI_API_KEY")
+                            }
+                        },
+                        "schema": schema,
+                        "extraction_type": "schema",
+                        "instruction": "From the crawled content    tioned asset names along with their prices and change in price.",
+                    }
+                },
+                "word_count_threshold": 1
+            }
+        }
     }
 
     try:
         result = tester.submit_and_wait(request)
-        extracted = json.loads(result["result"]["extracted_content"])
+        extracted = json.loads(result["result"]["results"][0]["extracted_content"])
         print(f"Extracted {len(extracted)} model pricing entries")
-        print("Sample entry:", json.dumps(extracted[0], indent=2))
+        if extracted:
+            print("Sample entry:", json.dumps(extracted[0], indent=2))
         assert result["result"]["success"]
     except Exception as e:
         print(f"LLM extraction test failed (might be due to missing API key): {str(e)}")
@@ -271,23 +300,32 @@ def test_llm_with_ollama(tester: Crawl4AiTester):
 
     request = {
         "urls": ["https://www.nbcnews.com/business"],
-        "priority": 8,
-        "extraction_config": {
-            "type": "llm",
+        "browser_config": {"verbose": True},
+        "crawler_config": {
+            "type": "CrawlerRunConfig",
             "params": {
-                "provider": "ollama/llama2",
-                "schema": schema,
-                "extraction_type": "schema",
-                "instruction": "Extract the main article information including title, summary, and main topics.",
-            },
-        },
-        "extra": {"word_count_threshold": 1},
-        "crawler_params": {"verbose": True},
+                "extraction_strategy": {
+                    "type": "LLMExtractionStrategy",
+                    "params": {
+                        "llm_config": {
+                            "type": "LLMConfig",
+                            "params": {
+                                "provider": "ollama/llama3.2:latest",
+                            }
+                        },
+                        "schema": schema,
+                        "extraction_type": "schema",
+                        "instruction": "Extract the main article information including title, summary, and main topics.",
+                    }
+                },
+                "word_count_threshold": 1
+            }
+        }
     }
 
     try:
         result = tester.submit_and_wait(request)
-        extracted = json.loads(result["result"]["extracted_content"])
+        extracted = json.loads(result["result"]["results"][0]["extracted_content"])
         print("Extracted content:", json.dumps(extracted, indent=2))
         assert result["result"]["success"]
     except Exception as e:
@@ -298,23 +336,29 @@ def test_cosine_extraction(tester: Crawl4AiTester):
     print("\n=== Testing Cosine Extraction ===")
     request = {
         "urls": ["https://www.nbcnews.com/business"],
-        "priority": 8,
-        "extraction_config": {
-            "type": "cosine",
+        "browser_config": {},
+        "crawler_config": {
+            "type": "CrawlerRunConfig",
             "params": {
-                "semantic_filter": "business finance economy",
-                "word_count_threshold": 10,
-                "max_dist": 0.2,
-                "top_k": 3,
-            },
-        },
+                "extraction_strategy": {
+                    "type": "CosineStrategy",
+                    "params": {
+                        "semantic_filter": "business finance economy",
+                        "word_count_threshold": 10,
+                        "max_dist": 0.2,
+                        "top_k": 3,
+                    }
+                }
+            }
+        }
     }
 
     try:
         result = tester.submit_and_wait(request)
-        extracted = json.loads(result["result"]["extracted_content"])
+        extracted = json.loads(result["result"]["results"][0]["extracted_content"])
         print(f"Extracted {len(extracted)} text clusters")
-        print("First cluster tags:", extracted[0]["tags"])
+        if extracted:
+            print("First cluster tags:", extracted[0]["tags"])
         assert result["result"]["success"]
     except Exception as e:
         print(f"Cosine extraction test failed: {str(e)}")
@@ -324,19 +368,24 @@ def test_screenshot(tester: Crawl4AiTester):
     print("\n=== Testing Screenshot ===")
     request = {
         "urls": ["https://www.nbcnews.com/business"],
-        "priority": 5,
-        "screenshot": True,
-        "crawler_params": {"headless": True},
+        "browser_config": {"headless": True},
+        "crawler_config": {
+            "type": "CrawlerRunConfig",
+            "params": {
+                "screenshot": True
+            }
+        }
     }
 
     result = tester.submit_and_wait(request)
-    print("Screenshot captured:", bool(result["result"]["screenshot"]))
+    screenshot_data = result["result"]["results"][0]["screenshot"]
+    print("Screenshot captured:", bool(screenshot_data))
 
-    if result["result"]["screenshot"]:
+    if screenshot_data:
         # Save screenshot
-        screenshot_data = base64.b64decode(result["result"]["screenshot"])
+        screenshot_bytes = base64.b64decode(screenshot_data)
         with open("test_screenshot.jpg", "wb") as f:
-            f.write(screenshot_data)
+            f.write(screenshot_bytes)
         print("Screenshot saved as test_screenshot.jpg")
 
     assert result["result"]["success"]

tests/general/test_persistent_context.py

@@ -0,0 +1,43 @@
+import asyncio
+import os
+from crawl4ai.async_webcrawler import AsyncWebCrawler
+from crawl4ai.async_configs import BrowserConfig, CrawlerRunConfig, CacheMode
+
+# Simple concurrency test for persistent context page creation
+# Usage: python scripts/test_persistent_context.py
+
+URLS = [
+    # "https://example.com",
+    "https://httpbin.org/html",
+    "https://www.python.org/",
+    "https://www.rust-lang.org/",
+]
+
+async def main():
+    profile_dir = os.path.join(os.path.expanduser("~"), ".crawl4ai", "profiles", "test-persistent-profile")
+    os.makedirs(profile_dir, exist_ok=True)
+
+    browser_config = BrowserConfig(
+        browser_type="chromium",
+        headless=True,
+        use_persistent_context=True,
+        user_data_dir=profile_dir,
+        use_managed_browser=True,
+        verbose=True,
+    )
+
+    run_cfg = CrawlerRunConfig(
+        cache_mode=CacheMode.BYPASS,
+        stream=False,
+        verbose=True,
+    )
+
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        results = await crawler.arun_many(URLS, config=run_cfg)
+        for r in results:
+            print(r.url, r.success, len(r.markdown.raw_markdown) if r.markdown else 0)
+        # r = await crawler.arun(url=URLS[0], config=run_cfg)
+        # print(r.url, r.success, len(r.markdown.raw_markdown) if r.markdown else 0)
+
+if __name__ == "__main__":
+    asyncio.run(main())

tests/profiler/test_keyboard_handle.py

@@ -0,0 +1,55 @@
+import sys
+import pytest
+import asyncio
+from unittest.mock import patch, MagicMock
+from crawl4ai.browser_profiler import BrowserProfiler
+
+@pytest.mark.asyncio
+@pytest.mark.skipif(sys.platform != "win32", reason="Windows-specific msvcrt test")
+async def test_keyboard_input_handling():
+    # Mock sequence of keystrokes: arrow key followed by 'q'
+    mock_keys = [b'\x00K', b'q']
+    mock_kbhit = MagicMock(side_effect=[True, True, False])
+    mock_getch = MagicMock(side_effect=mock_keys)
+
+    with patch('msvcrt.kbhit', mock_kbhit), patch('msvcrt.getch', mock_getch):
+        # profiler = BrowserProfiler()
+        user_done_event = asyncio.Event()
+        
+        # Create a local async function to simulate the keyboard input handling
+        async def test_listen_for_quit_command():
+            if sys.platform == "win32":
+                while True:
+                    try:
+                        if mock_kbhit():
+                            raw = mock_getch()
+                            try:
+                                key = raw.decode("utf-8")
+                            except UnicodeDecodeError:
+                                continue
+
+                            if len(key) != 1 or not key.isprintable():
+                                continue
+
+                            if key.lower() == "q":
+                                user_done_event.set()
+                                return
+
+                        await asyncio.sleep(0.1)
+                    except Exception as e:
+                        continue
+
+        # Run the listener
+        listener_task = asyncio.create_task(test_listen_for_quit_command())
+        
+        # Wait for the event to be set
+        try:
+            await asyncio.wait_for(user_done_event.wait(), timeout=1.0)
+            assert user_done_event.is_set()
+        finally:
+            if not listener_task.done():
+                listener_task.cancel()
+                try:
+                    await listener_task
+                except asyncio.CancelledError:
+                    pass

tests/proxy/test_proxy_config.py

@@ -0,0 +1,582 @@
+"""
+Comprehensive test suite for ProxyConfig in different forms:
+1. String form (ip:port:username:password)
+2. Dict form (dictionary with keys)
+3. Object form (ProxyConfig instance)
+4. Environment variable form (from env vars)
+
+Tests cover all possible scenarios and edge cases using pytest.
+"""
+
+import asyncio
+import os
+import pytest
+import tempfile
+from unittest.mock import patch
+
+from crawl4ai import AsyncWebCrawler, BrowserConfig
+from crawl4ai.async_configs import CrawlerRunConfig, ProxyConfig
+from crawl4ai.cache_context import CacheMode
+
+
+class TestProxyConfig:
+    """Comprehensive test suite for ProxyConfig functionality."""
+    
+    # Test data for different scenarios
+    # get free proxy server from from webshare.io https://www.webshare.io/?referral_code=3sqog0y1fvsl
+    TEST_PROXY_DATA = {
+        "server": "",
+        "username": "", 
+        "password": "",
+        "ip": ""
+    }
+    
+    def setup_method(self):
+        """Setup for each test method."""
+        self.test_url = "https://httpbin.org/ip"  # Use httpbin for testing
+        
+    # ==================== OBJECT FORM TESTS ====================
+    
+    def test_proxy_config_object_creation_basic(self):
+        """Test basic ProxyConfig object creation."""
+        proxy = ProxyConfig(server="127.0.0.1:8080")
+        assert proxy.server == "127.0.0.1:8080"
+        assert proxy.username is None
+        assert proxy.password is None
+        assert proxy.ip == "127.0.0.1"  # Should auto-extract IP
+        
+    def test_proxy_config_object_creation_full(self):
+        """Test ProxyConfig object creation with all parameters."""
+        proxy = ProxyConfig(
+            server=f"http://{self.TEST_PROXY_DATA['server']}",
+            username=self.TEST_PROXY_DATA['username'],
+            password=self.TEST_PROXY_DATA['password'],
+            ip=self.TEST_PROXY_DATA['ip']
+        )
+        assert proxy.server == f"http://{self.TEST_PROXY_DATA['server']}"
+        assert proxy.username == self.TEST_PROXY_DATA['username']
+        assert proxy.password == self.TEST_PROXY_DATA['password']
+        assert proxy.ip == self.TEST_PROXY_DATA['ip']
+        
+    def test_proxy_config_object_ip_extraction(self):
+        """Test automatic IP extraction from server URL."""
+        test_cases = [
+            ("http://192.168.1.1:8080", "192.168.1.1"),
+            ("https://10.0.0.1:3128", "10.0.0.1"),
+            ("192.168.1.100:8080", "192.168.1.100"),
+            ("proxy.example.com:8080", "proxy.example.com"),
+        ]
+        
+        for server, expected_ip in test_cases:
+            proxy = ProxyConfig(server=server)
+            assert proxy.ip == expected_ip, f"Failed for server: {server}"
+            
+    def test_proxy_config_object_invalid_server(self):
+        """Test ProxyConfig with invalid server formats."""
+        # Should not raise exception but may not extract IP properly
+        proxy = ProxyConfig(server="invalid-format")
+        assert proxy.server == "invalid-format"
+        # IP extraction might fail but object should still be created
+        
+    # ==================== DICT FORM TESTS ====================
+    
+    def test_proxy_config_from_dict_basic(self):
+        """Test creating ProxyConfig from basic dictionary."""
+        proxy_dict = {"server": "127.0.0.1:8080"}
+        proxy = ProxyConfig.from_dict(proxy_dict)
+        assert proxy.server == "127.0.0.1:8080"
+        assert proxy.username is None
+        assert proxy.password is None
+        
+    def test_proxy_config_from_dict_full(self):
+        """Test creating ProxyConfig from complete dictionary."""
+        proxy_dict = {
+            "server": f"http://{self.TEST_PROXY_DATA['server']}",
+            "username": self.TEST_PROXY_DATA['username'],
+            "password": self.TEST_PROXY_DATA['password'],
+            "ip": self.TEST_PROXY_DATA['ip']
+        }
+        proxy = ProxyConfig.from_dict(proxy_dict)
+        assert proxy.server == proxy_dict["server"]
+        assert proxy.username == proxy_dict["username"]
+        assert proxy.password == proxy_dict["password"]
+        assert proxy.ip == proxy_dict["ip"]
+        
+    def test_proxy_config_from_dict_missing_keys(self):
+        """Test creating ProxyConfig from dictionary with missing keys."""
+        proxy_dict = {"server": "127.0.0.1:8080", "username": "user"}
+        proxy = ProxyConfig.from_dict(proxy_dict)
+        assert proxy.server == "127.0.0.1:8080"
+        assert proxy.username == "user"
+        assert proxy.password is None
+        assert proxy.ip == "127.0.0.1"  # Should auto-extract
+        
+    def test_proxy_config_from_dict_empty(self):
+        """Test creating ProxyConfig from empty dictionary."""
+        proxy_dict = {}
+        proxy = ProxyConfig.from_dict(proxy_dict)
+        assert proxy.server is None
+        assert proxy.username is None
+        assert proxy.password is None
+        assert proxy.ip is None
+        
+    def test_proxy_config_from_dict_none_values(self):
+        """Test creating ProxyConfig from dictionary with None values."""
+        proxy_dict = {
+            "server": "127.0.0.1:8080",
+            "username": None,
+            "password": None,
+            "ip": None
+        }
+        proxy = ProxyConfig.from_dict(proxy_dict)
+        assert proxy.server == "127.0.0.1:8080"
+        assert proxy.username is None
+        assert proxy.password is None
+        assert proxy.ip == "127.0.0.1"  # Should auto-extract despite None
+        
+    # ==================== STRING FORM TESTS ====================
+    
+    def test_proxy_config_from_string_full_format(self):
+        """Test creating ProxyConfig from full string format (ip:port:username:password)."""
+        proxy_str = f"{self.TEST_PROXY_DATA['ip']}:6114:{self.TEST_PROXY_DATA['username']}:{self.TEST_PROXY_DATA['password']}"
+        proxy = ProxyConfig.from_string(proxy_str)
+        assert proxy.server == f"http://{self.TEST_PROXY_DATA['ip']}:6114"
+        assert proxy.username == self.TEST_PROXY_DATA['username']
+        assert proxy.password == self.TEST_PROXY_DATA['password']
+        assert proxy.ip == self.TEST_PROXY_DATA['ip']
+        
+    def test_proxy_config_from_string_ip_port_only(self):
+        """Test creating ProxyConfig from string with only ip:port."""
+        proxy_str = "192.168.1.1:8080"
+        proxy = ProxyConfig.from_string(proxy_str)
+        assert proxy.server == "http://192.168.1.1:8080"
+        assert proxy.username is None
+        assert proxy.password is None
+        assert proxy.ip == "192.168.1.1"
+        
+    def test_proxy_config_from_string_invalid_format(self):
+        """Test creating ProxyConfig from invalid string formats."""
+        invalid_formats = [
+            "invalid",
+            "ip:port:user",  # Missing password (3 parts)
+            "ip:port:user:pass:extra",  # Too many parts (5 parts)
+            "",
+            "::",  # Empty parts but 3 total (invalid)
+            "::::",  # Empty parts but 5 total (invalid)
+        ]
+        
+        for proxy_str in invalid_formats:
+            with pytest.raises(ValueError, match="Invalid proxy string format"):
+                ProxyConfig.from_string(proxy_str)
+                
+    def test_proxy_config_from_string_edge_cases_that_work(self):
+        """Test string formats that should work but might be edge cases."""
+        # These cases actually work as valid formats
+        edge_cases = [
+            (":", "http://:", ""),  # ip:port format with empty values
+            (":::", "http://:", ""),  # ip:port:user:pass format with empty values
+        ]
+        
+        for proxy_str, expected_server, expected_ip in edge_cases:
+            proxy = ProxyConfig.from_string(proxy_str)
+            assert proxy.server == expected_server
+            assert proxy.ip == expected_ip
+                
+    def test_proxy_config_from_string_edge_cases(self):
+        """Test string parsing edge cases."""
+        # Test with different port numbers
+        proxy_str = "10.0.0.1:3128:user:pass"
+        proxy = ProxyConfig.from_string(proxy_str)
+        assert proxy.server == "http://10.0.0.1:3128"
+        
+        # Test with special characters in credentials
+        proxy_str = "10.0.0.1:8080:user@domain:pass:word"
+        with pytest.raises(ValueError):  # Should fail due to extra colon in password
+            ProxyConfig.from_string(proxy_str)
+            
+    # ==================== ENVIRONMENT VARIABLE TESTS ====================
+    
+    def test_proxy_config_from_env_single_proxy(self):
+        """Test loading single proxy from environment variable."""
+        proxy_str = f"{self.TEST_PROXY_DATA['ip']}:6114:{self.TEST_PROXY_DATA['username']}:{self.TEST_PROXY_DATA['password']}"
+        
+        with patch.dict(os.environ, {'TEST_PROXIES': proxy_str}):
+            proxies = ProxyConfig.from_env('TEST_PROXIES')
+            assert len(proxies) == 1
+            proxy = proxies[0]
+            assert proxy.ip == self.TEST_PROXY_DATA['ip']
+            assert proxy.username == self.TEST_PROXY_DATA['username']
+            assert proxy.password == self.TEST_PROXY_DATA['password']
+            
+    def test_proxy_config_from_env_multiple_proxies(self):
+        """Test loading multiple proxies from environment variable."""
+        proxy_list = [
+            "192.168.1.1:8080:user1:pass1",
+            "192.168.1.2:8080:user2:pass2",
+            "10.0.0.1:3128"  # No auth
+        ]
+        proxy_str = ",".join(proxy_list)
+        
+        with patch.dict(os.environ, {'TEST_PROXIES': proxy_str}):
+            proxies = ProxyConfig.from_env('TEST_PROXIES')
+            assert len(proxies) == 3
+            
+            # Check first proxy
+            assert proxies[0].ip == "192.168.1.1"
+            assert proxies[0].username == "user1"
+            assert proxies[0].password == "pass1"
+            
+            # Check second proxy
+            assert proxies[1].ip == "192.168.1.2"
+            assert proxies[1].username == "user2"
+            assert proxies[1].password == "pass2"
+            
+            # Check third proxy (no auth)
+            assert proxies[2].ip == "10.0.0.1"
+            assert proxies[2].username is None
+            assert proxies[2].password is None
+            
+    def test_proxy_config_from_env_empty_var(self):
+        """Test loading from empty environment variable."""
+        with patch.dict(os.environ, {'TEST_PROXIES': ''}):
+            proxies = ProxyConfig.from_env('TEST_PROXIES')
+            assert len(proxies) == 0
+            
+    def test_proxy_config_from_env_missing_var(self):
+        """Test loading from missing environment variable."""
+        # Ensure the env var doesn't exist
+        with patch.dict(os.environ, {}, clear=True):
+            proxies = ProxyConfig.from_env('NON_EXISTENT_VAR')
+            assert len(proxies) == 0
+            
+    def test_proxy_config_from_env_with_empty_entries(self):
+        """Test loading proxies with empty entries in the list."""
+        proxy_str = "192.168.1.1:8080:user:pass,,10.0.0.1:3128,"
+        
+        with patch.dict(os.environ, {'TEST_PROXIES': proxy_str}):
+            proxies = ProxyConfig.from_env('TEST_PROXIES')
+            assert len(proxies) == 2  # Empty entries should be skipped
+            assert proxies[0].ip == "192.168.1.1"
+            assert proxies[1].ip == "10.0.0.1"
+            
+    def test_proxy_config_from_env_with_invalid_entries(self):
+        """Test loading proxies with some invalid entries."""
+        proxy_str = "192.168.1.1:8080:user:pass,invalid_proxy,10.0.0.1:3128"
+        
+        with patch.dict(os.environ, {'TEST_PROXIES': proxy_str}):
+            # Should handle errors gracefully and return valid proxies
+            proxies = ProxyConfig.from_env('TEST_PROXIES')
+            # Depending on implementation, might return partial list or empty
+            # This tests error handling
+            assert isinstance(proxies, list)
+            
+    # ==================== SERIALIZATION TESTS ====================
+    
+    def test_proxy_config_to_dict(self):
+        """Test converting ProxyConfig to dictionary."""
+        proxy = ProxyConfig(
+            server=f"http://{self.TEST_PROXY_DATA['server']}",
+            username=self.TEST_PROXY_DATA['username'],
+            password=self.TEST_PROXY_DATA['password'],
+            ip=self.TEST_PROXY_DATA['ip']
+        )
+        
+        result_dict = proxy.to_dict()
+        expected = {
+            "server": f"http://{self.TEST_PROXY_DATA['server']}",
+            "username": self.TEST_PROXY_DATA['username'],
+            "password": self.TEST_PROXY_DATA['password'],
+            "ip": self.TEST_PROXY_DATA['ip']
+        }
+        assert result_dict == expected
+        
+    def test_proxy_config_clone(self):
+        """Test cloning ProxyConfig with modifications."""
+        original = ProxyConfig(
+            server="http://127.0.0.1:8080",
+            username="user",
+            password="pass"
+        )
+        
+        # Clone with modifications
+        cloned = original.clone(username="new_user", password="new_pass")
+        
+        # Original should be unchanged
+        assert original.username == "user"
+        assert original.password == "pass"
+        
+        # Clone should have new values
+        assert cloned.username == "new_user"
+        assert cloned.password == "new_pass"
+        assert cloned.server == original.server  # Unchanged value
+        
+    def test_proxy_config_roundtrip_serialization(self):
+        """Test that ProxyConfig can be serialized and deserialized without loss."""
+        original = ProxyConfig(
+            server=f"http://{self.TEST_PROXY_DATA['server']}",
+            username=self.TEST_PROXY_DATA['username'],
+            password=self.TEST_PROXY_DATA['password'],
+            ip=self.TEST_PROXY_DATA['ip']
+        )
+        
+        # Serialize to dict and back
+        serialized = original.to_dict()
+        deserialized = ProxyConfig.from_dict(serialized)
+        
+        assert deserialized.server == original.server
+        assert deserialized.username == original.username
+        assert deserialized.password == original.password
+        assert deserialized.ip == original.ip
+        
+    # ==================== INTEGRATION TESTS ====================
+    
+    @pytest.mark.asyncio
+    async def test_crawler_with_proxy_config_object(self):
+        """Test AsyncWebCrawler with ProxyConfig object."""
+        proxy_config = ProxyConfig(
+            server=f"http://{self.TEST_PROXY_DATA['server']}",
+            username=self.TEST_PROXY_DATA['username'],
+            password=self.TEST_PROXY_DATA['password']
+        )
+        
+        browser_config = BrowserConfig(headless=True)
+        
+        # Test that the crawler accepts the ProxyConfig object without errors
+        async with AsyncWebCrawler(config=browser_config) as crawler:
+            try:
+                # Note: This might fail due to actual proxy connection, but should not fail due to config issues
+                result = await crawler.arun(
+                    url=self.test_url,
+                    config=CrawlerRunConfig(
+                        cache_mode=CacheMode.BYPASS,
+                        proxy_config=proxy_config,
+                        page_timeout=10000  # Short timeout for testing
+                    )
+                )
+                # If we get here, proxy config was accepted
+                assert result is not None
+            except Exception as e:
+                # We expect connection errors with test proxies, but not config errors
+                error_msg = str(e).lower()
+                assert "attribute" not in error_msg, f"Config error: {e}"
+                assert "proxy_config" not in error_msg, f"Proxy config error: {e}"
+                
+    @pytest.mark.asyncio
+    async def test_crawler_with_proxy_config_dict(self):
+        """Test AsyncWebCrawler with ProxyConfig from dictionary."""
+        proxy_dict = {
+            "server": f"http://{self.TEST_PROXY_DATA['server']}",
+            "username": self.TEST_PROXY_DATA['username'],
+            "password": self.TEST_PROXY_DATA['password']
+        }
+        proxy_config = ProxyConfig.from_dict(proxy_dict)
+        
+        browser_config = BrowserConfig(headless=True)
+        
+        async with AsyncWebCrawler(config=browser_config) as crawler:
+            try:
+                result = await crawler.arun(
+                    url=self.test_url,
+                    config=CrawlerRunConfig(
+                        cache_mode=CacheMode.BYPASS,
+                        proxy_config=proxy_config,
+                        page_timeout=10000
+                    )
+                )
+                assert result is not None
+            except Exception as e:
+                error_msg = str(e).lower()
+                assert "attribute" not in error_msg, f"Config error: {e}"
+                
+    @pytest.mark.asyncio
+    async def test_crawler_with_proxy_config_from_string(self):
+        """Test AsyncWebCrawler with ProxyConfig from string."""
+        proxy_str = f"{self.TEST_PROXY_DATA['ip']}:6114:{self.TEST_PROXY_DATA['username']}:{self.TEST_PROXY_DATA['password']}"
+        proxy_config = ProxyConfig.from_string(proxy_str)
+        
+        browser_config = BrowserConfig(headless=True)
+        
+        async with AsyncWebCrawler(config=browser_config) as crawler:
+            try:
+                result = await crawler.arun(
+                    url=self.test_url,
+                    config=CrawlerRunConfig(
+                        cache_mode=CacheMode.BYPASS,
+                        proxy_config=proxy_config,
+                        page_timeout=10000
+                    )
+                )
+                assert result is not None
+            except Exception as e:
+                error_msg = str(e).lower()
+                assert "attribute" not in error_msg, f"Config error: {e}"
+                
+    # ==================== EDGE CASES AND ERROR HANDLING ====================
+    
+    def test_proxy_config_with_none_server(self):
+        """Test ProxyConfig behavior with None server."""
+        proxy = ProxyConfig(server=None)
+        assert proxy.server is None
+        assert proxy.ip is None  # Should not crash
+        
+    def test_proxy_config_with_empty_string_server(self):
+        """Test ProxyConfig behavior with empty string server."""
+        proxy = ProxyConfig(server="")
+        assert proxy.server == ""
+        assert proxy.ip is None or proxy.ip == ""
+        
+    def test_proxy_config_special_characters_in_credentials(self):
+        """Test ProxyConfig with special characters in username/password."""
+        special_chars_tests = [
+            ("user@domain.com", "pass!@#$%"),
+            ("user_123", "p@ssw0rd"),
+            ("user-test", "pass-word"),
+        ]
+        
+        for username, password in special_chars_tests:
+            proxy = ProxyConfig(
+                server="http://127.0.0.1:8080",
+                username=username,
+                password=password
+            )
+            assert proxy.username == username
+            assert proxy.password == password
+            
+    def test_proxy_config_unicode_handling(self):
+        """Test ProxyConfig with unicode characters."""
+        proxy = ProxyConfig(
+            server="http://127.0.0.1:8080",
+            username="ユーザー",  # Japanese characters
+            password="пароль"    # Cyrillic characters
+        )
+        assert proxy.username == "ユーザー"
+        assert proxy.password == "пароль"
+        
+    # ==================== PERFORMANCE TESTS ====================
+    
+    def test_proxy_config_creation_performance(self):
+        """Test that ProxyConfig creation is reasonably fast."""
+        import time
+        
+        start_time = time.time()
+        for i in range(1000):
+            proxy = ProxyConfig(
+                server=f"http://192.168.1.{i % 255}:8080",
+                username=f"user{i}",
+                password=f"pass{i}"
+            )
+        end_time = time.time()
+        
+        # Should be able to create 1000 configs in less than 1 second
+        assert (end_time - start_time) < 1.0
+        
+    def test_proxy_config_from_env_performance(self):
+        """Test that loading many proxies from env is reasonably fast."""
+        import time
+        
+        # Create a large list of proxy strings
+        proxy_list = [f"192.168.1.{i}:8080:user{i}:pass{i}" for i in range(100)]
+        proxy_str = ",".join(proxy_list)
+        
+        with patch.dict(os.environ, {'PERF_TEST_PROXIES': proxy_str}):
+            start_time = time.time()
+            proxies = ProxyConfig.from_env('PERF_TEST_PROXIES')
+            end_time = time.time()
+            
+            assert len(proxies) == 100
+            # Should be able to parse 100 proxies in less than 1 second
+            assert (end_time - start_time) < 1.0
+
+
+# ==================== STANDALONE TEST FUNCTIONS ====================
+
+@pytest.mark.asyncio
+async def test_dict_proxy():
+    """Original test function for dict proxy - kept for backward compatibility."""
+    proxy_config = {
+        "server": "23.95.150.145:6114", 
+        "username": "cfyswbwn",
+        "password": "1gs266hoqysi"
+    }
+    proxy_config_obj = ProxyConfig.from_dict(proxy_config)
+    
+    browser_config = BrowserConfig(headless=True)
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        try:
+            result = await crawler.arun(url="https://httpbin.org/ip", config=CrawlerRunConfig(
+                stream=False,
+                cache_mode=CacheMode.BYPASS,
+                proxy_config=proxy_config_obj,
+                page_timeout=10000
+            ))
+            print("Dict proxy test passed!")
+            print(result.markdown[:200] if result and result.markdown else "No result")
+        except Exception as e:
+            print(f"Dict proxy test error (expected): {e}")
+
+
+@pytest.mark.asyncio
+async def test_string_proxy():
+    """Test function for string proxy format."""
+    proxy_str = "23.95.150.145:6114:cfyswbwn:1gs266hoqysi"
+    proxy_config_obj = ProxyConfig.from_string(proxy_str)
+    
+    browser_config = BrowserConfig(headless=True)
+    async with AsyncWebCrawler(config=browser_config) as crawler:
+        try:
+            result = await crawler.arun(url="https://httpbin.org/ip", config=CrawlerRunConfig(
+                stream=False,
+                cache_mode=CacheMode.BYPASS,
+                proxy_config=proxy_config_obj,
+                page_timeout=10000
+            ))
+            print("String proxy test passed!")
+            print(result.markdown[:200] if result and result.markdown else "No result")
+        except Exception as e:
+            print(f"String proxy test error (expected): {e}")
+
+
+@pytest.mark.asyncio
+async def test_env_proxy():
+    """Test function for environment variable proxy."""
+    # Set environment variable
+    os.environ['TEST_PROXIES'] = "23.95.150.145:6114:cfyswbwn:1gs266hoqysi"
+    
+    proxies = ProxyConfig.from_env('TEST_PROXIES')
+    if proxies:
+        proxy_config_obj = proxies[0]  # Use first proxy
+        
+        browser_config = BrowserConfig(headless=True)
+        async with AsyncWebCrawler(config=browser_config) as crawler:
+            try:
+                result = await crawler.arun(url="https://httpbin.org/ip", config=CrawlerRunConfig(
+                    stream=False,
+                    cache_mode=CacheMode.BYPASS,
+                    proxy_config=proxy_config_obj,
+                    page_timeout=10000
+                ))
+                print("Environment proxy test passed!")
+                print(result.markdown[:200] if result and result.markdown else "No result")
+            except Exception as e:
+                print(f"Environment proxy test error (expected): {e}")
+    else:
+        print("No proxies loaded from environment")
+
+
+if __name__ == "__main__":
+    print("Running comprehensive ProxyConfig tests...")
+    print("=" * 50)
+    
+    # Run the standalone test functions
+    print("\n1. Testing dict proxy format...")
+    asyncio.run(test_dict_proxy())
+    
+    print("\n2. Testing string proxy format...")
+    asyncio.run(test_string_proxy())
+    
+    print("\n3. Testing environment variable proxy format...")
+    asyncio.run(test_env_proxy())
+    
+    print("\n" + "=" * 50)
+    print("To run the full pytest suite, use: pytest " + __file__)
+    print("=" * 50)

tests/test_arun_many.py

@@ -0,0 +1,42 @@
+"""
+Test example for multiple crawler configs feature
+"""
+import asyncio
+import sys
+from pathlib import Path
+
+# Add parent directory to path for imports
+sys.path.insert(0, str(Path(__file__).parent.parent))
+
+from crawl4ai import AsyncWebCrawler, CrawlerRunConfig, CacheMode
+from crawl4ai.processors.pdf import PDFContentScrapingStrategy
+
+
+async def test_run_many():
+    default_config = CrawlerRunConfig(
+        cache_mode=CacheMode.BYPASS,
+        # scraping_strategy=PDFContentScrapingStrategy()
+    )
+    
+    test_urls = [
+        # "https://blog.python.org/",  # Blog URL  
+        "https://www.python.org/",  # Generic HTTPS page
+        "https://www.kidocode.com/",  # Generic HTTPS page
+        "https://www.example.com/",  # Generic HTTPS page
+        # "https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf",
+    ]
+    
+    async with AsyncWebCrawler() as crawler:
+        # Single config - traditional usage still works
+        print("Test 1: Single config (backwards compatible)")
+        result = await crawler.arun_many(
+            urls=test_urls[:2],
+            config=default_config
+        )
+        print(f"Crawled {len(result)} URLs with single config\n")
+        for item in result:
+            print(f"  {item.url} -> {item.status_code}")
+        
+
+if __name__ == "__main__":
+    asyncio.run(test_run_many())

tests/test_config_matching_only.py

@@ -0,0 +1,131 @@
+"""
+Test only the config matching logic without running crawler
+"""
+import sys
+from pathlib import Path
+
+# Add parent directory to path for imports
+sys.path.insert(0, str(Path(__file__).parent.parent))
+
+from crawl4ai.async_configs import CrawlerRunConfig, MatchMode
+
+def test_all_matching_scenarios():
+    print("Testing CrawlerRunConfig.is_match() method")
+    print("=" * 50)
+    
+    # Test 1: Single string pattern
+    print("\n1. Single string pattern (glob style)")
+    config = CrawlerRunConfig(
+        url_matcher="*.pdf",
+        # For example we can set this => scraping_strategy=PDFContentScrapingStrategy()
+    )
+    test_urls = [
+        ("https://example.com/file.pdf", True),
+        ("https://example.com/doc.PDF", False),  # Case sensitive
+        ("https://example.com/file.txt", False),
+        ("file.pdf", True),
+    ]
+    for url, expected in test_urls:
+        result = config.is_match(url)
+        status = "✓" if result == expected else "✗"
+        print(f"  {status} {url} -> {result}")
+    
+    # Test 2: List of patterns with OR
+    print("\n2. List of patterns with OR (default)")
+    config = CrawlerRunConfig(
+        url_matcher=["*/article/*", "*/blog/*", "*.html"],
+        match_mode=MatchMode.OR
+    )
+    test_urls = [
+        ("https://example.com/article/news", True),
+        ("https://example.com/blog/post", True),
+        ("https://example.com/page.html", True),
+        ("https://example.com/page.php", False),
+    ]
+    for url, expected in test_urls:
+        result = config.is_match(url)
+        status = "✓" if result == expected else "✗"
+        print(f"  {status} {url} -> {result}")
+    
+    # Test 3: Custom function
+    print("\n3. Custom function matcher")
+    config = CrawlerRunConfig(
+        url_matcher=lambda url: 'api' in url and (url.endswith('.json') or url.endswith('.xml'))
+    )
+    test_urls = [
+        ("https://api.example.com/data.json", True),
+        ("https://api.example.com/data.xml", True),
+        ("https://api.example.com/data.html", False),
+        ("https://example.com/data.json", False),  # No 'api'
+    ]
+    for url, expected in test_urls:
+        result = config.is_match(url)
+        status = "✓" if result == expected else "✗"
+        print(f"  {status} {url} -> {result}")
+    
+    # Test 4: Mixed list with AND
+    print("\n4. Mixed patterns and functions with AND")
+    config = CrawlerRunConfig(
+        url_matcher=[
+            "https://*",  # Must be HTTPS
+            lambda url: '.com' in url,  # Must have .com
+            lambda url: len(url) < 50  # Must be short
+        ],
+        match_mode=MatchMode.AND
+    )
+    test_urls = [
+        ("https://example.com/page", True),
+        ("http://example.com/page", False),  # Not HTTPS
+        ("https://example.org/page", False),  # No .com
+        ("https://example.com/" + "x" * 50, False),  # Too long
+    ]
+    for url, expected in test_urls:
+        result = config.is_match(url)
+        status = "✓" if result == expected else "✗"
+        print(f"  {status} {url} -> {result}")
+    
+    # Test 5: Complex real-world scenario
+    print("\n5. Complex pattern combinations")
+    config = CrawlerRunConfig(
+        url_matcher=[
+            "*/api/v[0-9]/*",  # API versioned endpoints
+            lambda url: 'graphql' in url,  # GraphQL endpoints
+            "*.json"  # JSON files
+        ],
+        match_mode=MatchMode.OR
+    )
+    test_urls = [
+        ("https://example.com/api/v1/users", True),
+        ("https://example.com/api/v2/posts", True),
+        ("https://example.com/graphql", True),
+        ("https://example.com/data.json", True),
+        ("https://example.com/api/users", False),  # No version
+    ]
+    for url, expected in test_urls:
+        result = config.is_match(url)
+        status = "✓" if result == expected else "✗"
+        print(f"  {status} {url} -> {result}")
+    
+    # Test 6: Edge cases
+    print("\n6. Edge cases")
+    
+    # No matcher
+    config = CrawlerRunConfig()
+    result = config.is_match("https://example.com")
+    print(f"  {'✓' if not result else '✗'} No matcher -> {result}")
+    
+    # Empty list
+    config = CrawlerRunConfig(url_matcher=[])
+    result = config.is_match("https://example.com")
+    print(f"  {'✓' if not result else '✗'} Empty list -> {result}")
+    
+    # None in list (should be skipped)
+    config = CrawlerRunConfig(url_matcher=["*.pdf", None, "*.doc"])
+    result = config.is_match("test.pdf")
+    print(f"  {'✓' if result else '✗'} List with None -> {result}")
+    
+    print("\n" + "=" * 50)
+    print("All matching tests completed!")
+
+if __name__ == "__main__":
+    test_all_matching_scenarios()

tests/test_config_selection.py

@@ -0,0 +1,87 @@
+"""
+Test config selection logic in dispatchers
+"""
+import asyncio
+import sys
+from pathlib import Path
+from unittest.mock import AsyncMock, MagicMock
+
+# Add parent directory to path for imports
+sys.path.insert(0, str(Path(__file__).parent.parent))
+
+from crawl4ai.async_configs import CrawlerRunConfig, MatchMode
+from crawl4ai.async_dispatcher import BaseDispatcher, MemoryAdaptiveDispatcher
+
+class TestDispatcher(BaseDispatcher):
+    """Simple test dispatcher to verify config selection"""
+    
+    async def crawl_url(self, url, config, task_id, **kwargs):
+        # Just return which config was selected
+        selected = self.select_config(url, config)
+        return {"url": url, "config_id": id(selected)}
+    
+    async def run_urls(self, urls, crawler, config):
+        results = []
+        for url in urls:
+            result = await self.crawl_url(url, config, "test")
+            results.append(result)
+        return results
+
+async def test_dispatcher_config_selection():
+    print("Testing dispatcher config selection")
+    print("=" * 50)
+    
+    # Create test configs with different matchers
+    pdf_config = CrawlerRunConfig(url_matcher="*.pdf")
+    api_config = CrawlerRunConfig(url_matcher=lambda url: 'api' in url)
+    default_config = CrawlerRunConfig()  # No matcher
+    
+    configs = [pdf_config, api_config, default_config]
+    
+    # Create test dispatcher
+    dispatcher = TestDispatcher()
+    
+    # Test single config
+    print("\nTest 1: Single config")
+    result = await dispatcher.crawl_url("https://example.com/file.pdf", pdf_config, "test1")
+    assert result["config_id"] == id(pdf_config)
+    print("✓ Single config works")
+    
+    # Test config list selection
+    print("\nTest 2: Config list selection")
+    test_cases = [
+        ("https://example.com/file.pdf", id(pdf_config)),
+        ("https://api.example.com/data", id(api_config)),
+        ("https://example.com/page", id(configs[0])),  # No match, uses first
+    ]
+    
+    for url, expected_id in test_cases:
+        result = await dispatcher.crawl_url(url, configs, "test")
+        assert result["config_id"] == expected_id, f"URL {url} got wrong config"
+        print(f"✓ {url} -> correct config selected")
+    
+    # Test with MemoryAdaptiveDispatcher
+    print("\nTest 3: MemoryAdaptiveDispatcher config selection")
+    mem_dispatcher = MemoryAdaptiveDispatcher()
+    
+    # Test select_config method directly
+    selected = mem_dispatcher.select_config("https://example.com/doc.pdf", configs)
+    assert selected == pdf_config
+    print("✓ MemoryAdaptiveDispatcher.select_config works")
+    
+    # Test empty config list
+    print("\nTest 4: Edge cases")
+    selected = mem_dispatcher.select_config("https://example.com", [])
+    assert isinstance(selected, CrawlerRunConfig)  # Should return default
+    print("✓ Empty config list returns default config")
+    
+    # Test None config
+    selected = mem_dispatcher.select_config("https://example.com", None)
+    assert isinstance(selected, CrawlerRunConfig)  # Should return default
+    print("✓ None config returns default config")
+    
+    print("\n" + "=" * 50)
+    print("All dispatcher tests passed! ✓")
+
+if __name__ == "__main__":
+    asyncio.run(test_dispatcher_config_selection())

tests/test_docker_api_with_llm_provider.py

@@ -0,0 +1,122 @@
+#!/usr/bin/env python3
+"""Test script to verify Docker API with LLM provider configuration."""
+
+import requests
+import json
+import time
+
+BASE_URL = "http://localhost:11235"
+
+def test_health():
+    """Test health endpoint."""
+    print("1. Testing health endpoint...")
+    response = requests.get(f"{BASE_URL}/health")
+    print(f"   Status: {response.status_code}")
+    print(f"   Response: {response.json()}")
+    print()
+
+def test_schema():
+    """Test schema endpoint to see configuration."""
+    print("2. Testing schema endpoint...")
+    response = requests.get(f"{BASE_URL}/schema")
+    print(f"   Status: {response.status_code}")
+    # Print only browser config to keep output concise
+    print(f"   Browser config keys: {list(response.json().get('browser', {}).keys())[:5]}...")
+    print()
+
+def test_markdown_with_llm_filter():
+    """Test markdown endpoint with LLM filter (should use configured provider)."""
+    print("3. Testing markdown endpoint with LLM filter...")
+    print("   This should use the Groq provider from LLM_PROVIDER env var")
+    
+    # Note: This will fail with dummy API keys, but we can see if it tries to use Groq
+    payload = {
+        "url": "https://httpbin.org/html",
+        "f": "llm",
+        "q": "Extract the main content"
+    }
+    
+    response = requests.post(f"{BASE_URL}/md", json=payload)
+    print(f"   Status: {response.status_code}")
+    
+    if response.status_code != 200:
+        print(f"   Error: {response.text[:200]}...")
+    else:
+        print(f"   Success! Markdown length: {len(response.json().get('markdown', ''))} chars")
+    print()
+
+def test_markdown_with_provider_override():
+    """Test markdown endpoint with provider override in request."""
+    print("4. Testing markdown endpoint with provider override...")
+    print("   This should use OpenAI provider from request parameter")
+    
+    payload = {
+        "url": "https://httpbin.org/html",
+        "f": "llm",
+        "q": "Extract the main content",
+        "provider": "openai/gpt-4"  # Override to use OpenAI
+    }
+    
+    response = requests.post(f"{BASE_URL}/md", json=payload)
+    print(f"   Status: {response.status_code}")
+    
+    if response.status_code != 200:
+        print(f"   Error: {response.text[:200]}...")
+    else:
+        print(f"   Success! Markdown length: {len(response.json().get('markdown', ''))} chars")
+    print()
+
+def test_simple_crawl():
+    """Test simple crawl without LLM."""
+    print("5. Testing simple crawl (no LLM required)...")
+    
+    payload = {
+        "urls": ["https://httpbin.org/html"],
+        "browser_config": {
+            "type": "BrowserConfig",
+            "params": {"headless": True}
+        },
+        "crawler_config": {
+            "type": "CrawlerRunConfig",
+            "params": {"cache_mode": "bypass"}
+        }
+    }
+    
+    response = requests.post(f"{BASE_URL}/crawl", json=payload)
+    print(f"   Status: {response.status_code}")
+    
+    if response.status_code == 200:
+        result = response.json()
+        print(f"   Success: {result.get('success')}")
+        print(f"   Results count: {len(result.get('results', []))}")
+        if result.get('results'):
+            print(f"   First result success: {result['results'][0].get('success')}")
+    else:
+        print(f"   Error: {response.text[:200]}...")
+    print()
+
+def test_playground():
+    """Test if playground is accessible."""
+    print("6. Testing playground interface...")
+    response = requests.get(f"{BASE_URL}/playground")
+    print(f"   Status: {response.status_code}")
+    print(f"   Content-Type: {response.headers.get('content-type')}")
+    print()
+
+if __name__ == "__main__":
+    print("=== Crawl4AI Docker API Tests ===\n")
+    print(f"Testing API at {BASE_URL}\n")
+    
+    # Wait a bit for server to be fully ready
+    time.sleep(2)
+    
+    test_health()
+    test_schema()
+    test_simple_crawl()
+    test_playground()
+    
+    print("\nTesting LLM functionality (these may fail with dummy API keys):\n")
+    test_markdown_with_llm_filter()
+    test_markdown_with_provider_override()
+    
+    print("\nTests completed!")

tests/test_memory_macos.py

@@ -0,0 +1,71 @@
+#!/usr/bin/env python3
+"""Test script to verify macOS memory calculation accuracy."""
+
+import psutil
+import platform
+import time
+from crawl4ai.memory_utils import get_true_memory_usage_percent, get_memory_stats, get_true_available_memory_gb
+
+
+def test_memory_calculation():
+    """Test and compare memory calculations."""
+    print(f"Platform: {platform.system()}")
+    print(f"Python version: {platform.python_version()}")
+    print("-" * 60)
+    
+    # Get psutil's view
+    vm = psutil.virtual_memory()
+    psutil_percent = vm.percent
+    psutil_available_gb = vm.available / (1024**3)
+    total_gb = vm.total / (1024**3)
+    
+    # Get our corrected view
+    true_percent = get_true_memory_usage_percent()
+    true_available_gb = get_true_available_memory_gb()
+    true_percent_calc, available_calc, total_calc = get_memory_stats()
+    
+    print("Memory Statistics Comparison:")
+    print(f"Total Memory: {total_gb:.2f} GB")
+    print()
+    
+    print("PSUtil (Standard) Calculation:")
+    print(f"  - Memory Used: {psutil_percent:.1f}%")
+    print(f"  - Available: {psutil_available_gb:.2f} GB")
+    print()
+    
+    print("Platform-Aware Calculation:")
+    print(f"  - Memory Used: {true_percent:.1f}%")
+    print(f"  - Available: {true_available_gb:.2f} GB")
+    print(f"  - Difference: {true_available_gb - psutil_available_gb:.2f} GB of reclaimable memory")
+    print()
+    
+    # Show the impact on dispatcher behavior
+    print("Impact on MemoryAdaptiveDispatcher:")
+    thresholds = {
+        "Normal": 90.0,
+        "Critical": 95.0,
+        "Recovery": 85.0
+    }
+    
+    for name, threshold in thresholds.items():
+        psutil_triggered = psutil_percent >= threshold
+        true_triggered = true_percent >= threshold
+        print(f"  - {name} Threshold ({threshold}%):")
+        print(f"    PSUtil: {'TRIGGERED' if psutil_triggered else 'OK'}")
+        print(f"    Platform-Aware: {'TRIGGERED' if true_triggered else 'OK'}")
+        if psutil_triggered != true_triggered:
+            print(f"    → Difference: Platform-aware prevents false {'pressure' if psutil_triggered else 'recovery'}")
+    print()
+    
+    # Monitor for a few seconds
+    print("Monitoring memory for 10 seconds...")
+    for i in range(10):
+        vm = psutil.virtual_memory()
+        true_pct = get_true_memory_usage_percent()
+        print(f"  {i+1}s - PSUtil: {vm.percent:.1f}% | Platform-Aware: {true_pct:.1f}%", end="\r")
+        time.sleep(1)
+    print("\n")
+
+
+if __name__ == "__main__":
+    test_memory_calculation()

tests/test_multi_config.py

@@ -0,0 +1,117 @@
+"""
+Test example for multiple crawler configs feature
+"""
+import asyncio
+import sys
+from pathlib import Path
+
+# Add parent directory to path for imports
+sys.path.insert(0, str(Path(__file__).parent.parent))
+
+from crawl4ai import AsyncWebCrawler, CrawlerRunConfig, MatchMode, CacheMode
+
+async def test_multi_config():
+    # Create different configs for different URL patterns
+    
+    # Config for PDF files
+    pdf_config = CrawlerRunConfig(
+        url_matcher="*.pdf",
+    )
+    
+    # Config for articles (using multiple patterns with OR logic)
+    article_config = CrawlerRunConfig(
+        url_matcher=["*/news/*", "*blog*", "*/article/*"],
+        match_mode=MatchMode.OR,
+        screenshot=True,
+    )
+    
+    # Config using custom matcher function
+    api_config = CrawlerRunConfig(
+        url_matcher=lambda url: 'api' in url or 'json' in url,
+    )
+    
+    # Config combining patterns and functions with AND logic
+    secure_docs_config = CrawlerRunConfig(
+        url_matcher=[
+            "*.doc*",  # Matches .doc, .docx
+            lambda url: url.startswith('https://')  # Must be HTTPS
+        ],
+        match_mode=MatchMode.AND,
+    )
+    
+    # Default config (no url_matcher means it won't match anything unless it's the fallback)
+    default_config = CrawlerRunConfig(
+        # cache_mode=CacheMode.BYPASS,
+    )
+    
+    # List of configs - order matters! First match wins
+    configs = [
+        pdf_config,
+        article_config, 
+        api_config,
+        secure_docs_config,
+        default_config  # Fallback
+    ]
+    
+    # Test URLs - using real URLs that exist
+    test_urls = [
+        "https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf",  # Real PDF
+        "https://www.bbc.com/news/articles/c5y3e3glnldo",  # News article
+        "https://blog.python.org/",  # Blog URL  
+        "https://api.github.com/users/github",  # GitHub API (returns JSON)
+        "https://httpbin.org/json",  # API endpoint that returns JSON
+        "https://www.python.org/",  # Generic HTTPS page
+        "http://info.cern.ch/",  # HTTP (not HTTPS) page
+        "https://example.com/",  # → Default config
+    ]
+    
+    # Test the matching logic
+    print("Config matching test:")
+    print("-" * 50)
+    for url in test_urls:
+        for i, config in enumerate(configs):
+            if config.is_match(url):
+                print(f"{url} -> Config {i} matches")
+                break
+        else:
+            print(f"{url} -> No match, will use fallback (first config)")
+    
+    print("\n" + "=" * 50 + "\n")
+    
+    # Now test with actual crawler
+    async with AsyncWebCrawler() as crawler:
+        # Single config - traditional usage still works
+        print("Test 1: Single config (backwards compatible)")
+        result = await crawler.arun_many(
+            urls=["https://www.python.org/"],
+            config=default_config
+        )
+        print(f"Crawled {len(result)} URLs with single config\n")
+        
+        # Multiple configs - new feature
+        print("Test 2: Multiple configs")
+        # Just test with 2 URLs to avoid timeout
+        results = await crawler.arun_many(
+            urls=test_urls[:2],  # Just test first 2 URLs
+            config=configs  # Pass list of configs
+        )
+        print(f"Crawled {len(results)} URLs with multiple configs")
+        
+        # Using custom matcher inline
+        print("\nTest 3: Inline custom matcher")
+        custom_config = CrawlerRunConfig(
+            url_matcher=lambda url: len(url) > 50 and 'python' in url.lower(),
+            verbose=False
+        )
+        results = await crawler.arun_many(
+            urls=[
+                "https://docs.python.org/3/library/asyncio.html",  # Long URL with 'python'
+                "https://python.org/",  # Short URL with 'python' - won't match
+                "https://www.google.com/"  # No 'python' - won't match
+            ],
+            config=[custom_config, default_config]
+        )
+        print(f"Crawled {len(results)} URLs with custom matcher")
+
+if __name__ == "__main__":
+    asyncio.run(test_multi_config())