Skip to main content
Glama
MCP_QUICKSTART.mdโ€ข4.53 kB
# ๐Ÿค– WebClone MCP Server - 2-Minute Quick Start ## What is This? WebClone is now an **official MCP (Model Context Protocol) server** that lets AI agents like Claude download websites and files! --- ## Installation (One Time) ```bash # Install WebClone with MCP support make install-mcp ``` That's it! Installation complete. --- ## Usage ### Option 1: With Claude Desktop (Recommended) 1. **Add to Claude's config**: Edit `~/.config/claude/claude_desktop_config.json`: ```json { "mcpServers": { "webclone": { "command": "python", "args": ["/absolute/path/to/webclone/webclone-mcp.py"] } } } ``` ๐Ÿ’ก Replace `/absolute/path/to/webclone/` with your actual path! 2. **Restart Claude Desktop** 3. **Use it!** Ask Claude: ``` Clone the FastAPI documentation website ``` Claude will use WebClone to download it automatically! ๐ŸŽ‰ ### Option 2: Standalone Server ```bash # Run MCP server on stdio make mcp ``` --- ## Available Tools for AI Agents Your AI agent can now: | Tool | What It Does | |------|--------------| | **clone_website** | Download entire websites with all pages & assets | | **download_file** | Download a specific file or URL | | **save_authentication** | Guide for saving login cookies | | **list_saved_sessions** | List all saved authentication sessions | | **get_site_info** | Get website metadata without downloading | --- ## Example Conversations with Claude ### Example 1: Simple Clone ``` You: Clone https://example.com Claude: I'll clone that website for you. [Uses clone_website tool] โœ… Success! Cloned 5 pages, 12 assets, 2.3 MB total. Output directory: ./website_mirror ``` ### Example 2: With Limits ``` You: Clone https://docs.python.org but limit it to 50 pages Claude: I'll clone the Python docs with a 50-page limit. [Uses clone_website with max_pages=50] โœ… Cloned 50 pages, 234 assets, 15.8 MB. ``` ### Example 3: Download Specific File ``` You: Download the PDF from https://example.com/report.pdf Claude: I'll download that PDF for you. [Uses download_file tool] โœ… Downloaded to ./downloads/ ``` --- ## Authentication (Optional) For sites requiring login (Google Drive, GitHub, etc.): 1. **Create session using GUI**: ```bash make gui ``` 2. **Go to "๐Ÿ” Authentication"** tab 3. **Save your session** (e.g., "google_drive") 4. **Use in Claude**: ``` Clone my Google Drive folder using the google_drive session ``` Claude will automatically use your saved cookies! ๐Ÿ” --- ## Integration with CrewAI ```python from crewai import Agent, Task, Crew # WebClone tool automatically available via MCP researcher = Agent( role='Web Archiver', goal='Clone documentation sites', tools=['clone_website'], # MCP tool verbose=True ) task = Task( description='Clone https://fastapi.tiangolo.com', agent=researcher ) crew = Crew(agents=[researcher], tasks=[task]) result = crew.kickoff() ``` --- ## Troubleshooting ### Claude can't see WebClone tools 1. โœ… Check config file path is correct 2. โœ… Use **absolute** paths (not relative) 3. โœ… Restart Claude Desktop completely 4. โœ… Ask Claude: "What tools do you have?" ### MCP server won't start ```bash # Reinstall make install-mcp # Test python webclone-mcp.py ``` ### Downloads fail 1. โœ… Check URL is valid (`http://` or `https://`) 2. โœ… Check output directory permissions 3. โœ… For auth sites, create session first --- ## Advanced Usage ### Speed Optimization ``` Clone https://example.com with 20 workers and 50ms delay ``` ### Stealth Mode ``` Clone https://example.com slowly with 3 workers and 2 second delay ``` ### Get Info First ``` Get information about https://example.com before cloning ``` --- ## Full Documentation - **Complete MCP Guide**: `docs/MCP_GUIDE.md` - **Authentication Help**: `docs/AUTHENTICATION_GUIDE.md` - **CLI Reference**: `docs/CLI_GUIDE.md` - **GUI Guide**: `docs/GUI_GUIDE.md` --- ## What Makes WebClone MCP Special? โœ… **Official MCP Tool** - Designed for AI agents โœ… **5 Powerful Tools** - Clone, download, authenticate, inspect โœ… **Production Ready** - Battle-tested website cloning engine โœ… **Authentication Support** - Access protected sites โœ… **Stealth Mode** - Bypass bot detection โœ… **Fast & Async** - Concurrent downloads โœ… **CrewAI Compatible** - Works with all AI frameworks --- **That's it! AI agents can now clone websites! ๐Ÿš€** **Author**: Ruslan Magana | **Website**: ruslanmv.com

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/ruslanmv/webclone'

If you have feedback or need assistance with the MCP directory API, please join our Discord server