DocsFetcher
Fetches package documentation from various language ecosystems without requiring API keys.
📚 DocsFetcher MCP Server
An MCP server that fetches package documentation from multiple language ecosystems for LLMs like Claude without requiring API keys.
✨ Features
- 🌐 Supports multiple programming languages (JavaScript, Python, Java, .NET, Ruby, PHP, Rust, Go, Swift)
- 📦 Fetches documentation for packages by name or URL
- 🔍 Crawls documentation sites to extract comprehensive information
- 📄 Extracts README, API docs, code examples, and repository info
- 🧠 Provides structured data for LLM summarization
- 💬 Includes specialized prompts for documentation analysis
- 🔑 No API key required - works natively with Claude Desktop and Cursor IDE
🚀 Installation
Claude Desktop
- Open Claude Desktop → Settings → Developer
- Click "Edit Config" and add:
{
"mcpServers": {
"docsFetcher": {
"command": "npx",
"args": [
"-y",
"@smithery/cli@latest",
"run",
"@cdugo/mcp-get-docs",
"--config",
"'{}'"
]
}
}
}
Cursor IDE Configuration
- Open Cursor IDE → Settings → MCP -> Add New MCP Servier
- Add:
Name: docsFetcher
Command: npx -y @smithery/cli@latest run @cdugo/mcp-get-docs --config "{}"
Prerequisites
- 📋 Node.js 18 or later
🏃♂️ Running Locally
git clone https://github.com/cdugo/package-documentation-mcp
cd package-documentation-mcp
npm install
npm run build
Once installed, you can run the server locally with:
# From the project root directory
npm start
For development with auto-restart on file changes:
npm run dev
The server will start on the default port (usually 3000). You should see output like:
🚀 DocsFetcher MCP Server running!
📋 Ready to fetch documentation
To specify a custom port:
PORT=8080 npm start
🛠️ Available Tools
- fetch-url-docs: 🔗 Fetch docs from a specific URL
- fetch-package-docs: 📦 Fetch docs for a package with optional language specification
- fetch-library-docs: 🧠 Smart tool that works with either package name or URL
- fetch-multilingual-docs: 🌍 Fetch docs for a package across multiple language ecosystems
📝 Available Prompts
- summarize-library-docs: 📚 Create a comprehensive library summary
- explain-dependency-error: 🐛 Generate dependency error explanations
💡 Example Queries
Basic Library Information
- "What is Express.js and how do I use it?"
- "Tell me about the React library"
- "How do I use requests in Python?"
Multi-language Support
- "Show me documentation for lodash in JavaScript"
- "Compare pandas in Python and data.table in R"
Using Tools
- "@fetch-package-docs with packageName='express' and language='javascript'"
- "@fetch-package-docs with packageName='requests' and language='python'"
- "@fetch-multilingual-docs with packageName='http' and languages=['javascript', 'python', 'rust']"
Using Prompts
- "@summarize-library-docs with libraryName='express'"
- "@explain-dependency-error with packageName='dotenv'"
❓ Troubleshooting
Local Installation
- Server not showing up: ✅ Verify absolute path in configuration
- Connection errors: 🔄 Restart Claude Desktop or Cursor IDE
- Fetch failures: ⚠️ Some packages may have non-standard documentation
- Language support: 🌐 If a language isn't working, try using the package's direct URL
📄 License
MIT
Máy chủ liên quan
Scout Monitoring MCP
nhà tài trợPut performance and error data directly in the hands of your AI assistant.
Alpha Vantage MCP Server
nhà tài trợAccess financial market data: realtime & historical stock, ETF, options, forex, crypto, commodities, fundamentals, technical indicators, & more
MCP Shell
Execute secure shell commands from AI assistants and other MCP clients, with configurable security settings.
Figma MCP Server
Enables AI assistants to interact with Figma via WebSocket for reading data and design analysis.
Email MCP Server by Sidemail
Let AI agents write & manage your SaaS emails
DevContainer MCP Server
Manage DevContainer environments using natural language prompts in any MCP-compatible editor.
Remote MCP Server Authless
An example of a remote MCP server deployable on Cloudflare Workers without authentication.
MCP Server Starter Template
A starter template for building Model Context Protocol (MCP) servers, designed for UI libraries and component registries.
Vega-Lite
Generate visualizations from fetched data using the VegaLite format and renderer.
Geo Location Demo
Retrieves user geolocation information using EdgeOne Pages Functions and integrates it with large language models via MCP.
mcp-rubber-duck
Query multiple LLMs in parallel from AI coding tools — rubber duck debugging, but the ducks talk back.
Workspace-Qdrant-MCP
Code knowledge and metadata with live update, knowledge library, semantic and vector searches