Context7 MCP Clone
Provides up-to-date, version-specific documentation and code examples for software libraries directly to LLMs, enabling resolution of library identifiers and retrieval of relevant documentation with code snippets.
README
Context7 MCP Clone - Up-to-Date Code Documentation for LLMs
A production-ready clone of Context7 MCP that provides up-to-date, version-specific documentation and code examples directly to LLMs and AI code editors.
๐ฏ Project Status
Phase 2 Complete: Core MCP Server and Backend API ready for development!
Completed Components
- โ pnpm monorepo with workspace configuration
- โ Docker Compose development and production environments
- โ PostgreSQL database schema with migrations
- โ NestJS backend API with authentication
- โ Library and documentation services
- โ MCP server with both tools (resolve-library-id, get-library-docs)
- โ Stdio and HTTP transports for MCP protocol
In Progress
- ๐ Rate limiting with Redis
- ๐ Crawler engine with GitHub API and web scraping
- ๐ Markdown and HTML parsing with code extraction
- ๐ Next.js web UI with Grafana theme
๐ฆ Project Structure
atamai-mcp/
โโโ packages/
โ โโโ mcp-server/ # MCP Protocol Server
โ โโโ backend-api/ # NestJS REST API
โ โโโ crawler-engine/ # Documentation Crawler
โ โโโ web-ui/ # Next.js Frontend
โโโ docker/ # Container configurations
โโโ scripts/ # Utility scripts
โโโ docker-compose.yml # Production compose
โโโ docker-compose.dev.yml # Development compose
โโโ PLAN.md # Detailed implementation plan
โโโ TODO.md # Task checklist
โโโ README.md # This file
๐ Quick Start
Prerequisites
- Node.js 20+
- pnpm 8+
- Docker & Docker Compose
- PostgreSQL 16+ (or use docker-compose)
Development Setup
- Clone and install dependencies
cd /Users/atamai/develope/atamai-mcp
pnpm install
- Set up environment
cp .env.example .env
# Edit .env with your configuration
- Start development environment with Docker
docker-compose -f docker-compose.dev.yml up
This starts:
- PostgreSQL database on port 5432
- Redis cache on port 6379
- Backend API on port 5000
- MCP Server on port 3000
- Web UI on port 4000
- Run migrations
# Migrations run automatically when backend starts
# Or manually run:
cd packages/backend-api
pnpm run migration:run
๐ MCP Server
Tools Available
1. resolve-library-id
Converts a library name to Context7-compatible ID and returns ranked results.
Input:
{
"libraryName": "react"
}
Output:
{
"libraries": [
{
"id": "/facebook/react",
"name": "React",
"description": "A JavaScript library for building user interfaces",
"ecosystem": "javascript",
"stars": 220000,
"benchmarkScore": 98,
"reputation": "high",
"codeSnippets": 15420
}
],
"selected": "/facebook/react",
"reasoning": "Exact name match with highest benchmark score"
}
2. get-library-docs
Retrieves version-specific documentation and code examples.
Input:
{
"context7CompatibleLibraryID": "/facebook/react",
"topic": "hooks",
"page": 1,
"mode": "code"
}
Output:
{
"libraryId": "/facebook/react",
"library": {
"name": "React",
"full_name": "facebook/react",
"ecosystem": "javascript"
},
"version": "18.2.0",
"topic": "hooks",
"page": 1,
"totalPages": 5,
"documentation": [
{
"title": "Using the State Hook",
"type": "guide",
"content": "...",
"codeExamples": [
{
"language": "javascript",
"code": "const [count, setCount] = useState(0);",
"description": "Basic useState example"
}
],
"url": "https://react.dev/reference/react/useState"
}
]
}
Running MCP Server
Stdio Mode (for Claude Desktop, Cursor, etc.)
cd packages/mcp-server
pnpm start
# Or with hot reload:
pnpm dev
HTTP Mode
MCP_TRANSPORT=http MCP_SERVER_PORT=3000 pnpm start
# Access SSE endpoint at: http://localhost:3000/mcp/sse
๐ Backend API
Authentication
Register
POST /api/v1/auth/register
Content-Type: application/json
{
"email": "user@example.com",
"password": "SecurePassword123!"
}
Login
POST /api/v1/auth/login
Content-Type: application/json
{
"email": "user@example.com",
"password": "SecurePassword123!"
}
Libraries
Search Libraries
GET /api/v1/libraries?query=react&ecosystem=javascript&page=1&limit=20
Authorization: Bearer <jwt_token>
Get Library by ID
GET /api/v1/libraries/{id}
Authorization: Bearer <jwt_token>
Get Available Ecosystems
GET /api/v1/libraries/ecosystems
Authorization: Bearer <jwt_token>
Documentation
Get Documentation
GET /api/v1/docs/{libraryId}?topic=hooks&page=1&mode=code
Authorization: Bearer <jwt_token>
Search Documentation
GET /api/v1/docs/search/{query}
Authorization: Bearer <jwt_token>
Swagger Documentation
Once the backend is running, visit:
http://localhost:5000/docs
๐ Database Schema
Core Tables
- users - User accounts with tier system
- api_keys - API key management with rate limits
- libraries - Library metadata
- library_versions - Version tracking
- documentation_pages - Documentation content with full-text search
- code_examples - Code snippets
- crawl_jobs - Crawler job tracking
- api_usage - Usage analytics
Features
- Full-text search on documentation
- Trigram similarity for fuzzy matching
- JSONB metadata support
- Automatic tsvector updates
- Materialized views for popular libraries
๐ Deployment
Production Docker Compose
docker-compose up -d
Environment Variables
See .env.example for all available options:
DATABASE_URL- PostgreSQL connection stringREDIS_URL- Redis connection stringJWT_SECRET- JWT signing keyNODE_ENV- production/developmentAPI_PREFIX- API base path (default: /api/v1)
Kubernetes Deployment
Deploy manifests are in the plan for Phase 8.
๐ Development Roadmap
Phase 3: Crawler Engine (Weeks 5-6)
- [ ] BullMQ job queue setup
- [ ] GitHub crawler with Octokit
- [ ] Documentation site scraper
- [ ] Markdown/HTML parsers
- [ ] Code extraction engine
Phase 4: Web UI (Weeks 7-8)
- [ ] Next.js 15 setup with App Router
- [ ] Landing page with purple gradient theme
- [ ] Documentation browser
- [ ] Authentication pages
Phase 5: Dashboard & Admin (Weeks 9-10)
- [ ] User dashboard
- [ ] API key management
- [ ] Admin panel
- [ ] Analytics and charts
Phase 6-10: Completion & Launch
- [ ] Data seeding
- [ ] Testing & optimization
- [ ] Production deployment
- [ ] Beta & public launch
๐งช Testing
# Run tests for all packages
pnpm test
# Run tests with coverage
pnpm test:cov
# Watch mode
pnpm test:watch
๐ Code Style
# Format code
pnpm format
# Check formatting
pnpm format:check
# Lint
pnpm lint
# Type check
pnpm type-check
๐ Debugging
Backend API
DEBUG=* pnpm dev
MCP Server
DEBUG=* pnpm dev
Docker Logs
docker-compose -f docker-compose.dev.yml logs -f [service-name]
๐ Documentation
- PLAN.md - Comprehensive implementation plan with architecture
- TODO.md - Detailed task checklist for all phases
- API Docs - Available at
http://localhost:5000/docs(Swagger)
๐ค Contributing
This is a solo development project. Please follow:
- TypeScript strict mode
- ESLint configuration
- Prettier formatting
- 80%+ test coverage
๐ License
MIT
๐ฏ Key Metrics (Target)
- Performance: MCP response <200ms (p95)
- Search: <100ms (p95)
- Uptime: 99.9%
- Coverage: 100+ libraries by month 6
- Users: 1000+ registered by month 3
Next Steps
- Run
docker-compose -f docker-compose.dev.yml up - Wait for migrations to complete
- Visit http://localhost:5000/docs for API documentation
- Start using the MCP server with Claude Desktop or Cursor
- Reference PLAN.md for detailed architecture and TODO.md for task tracking
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
E2B
Using MCP to run code via e2b.