mcpdir
The largest open MCP server directory — 8,000+ community-driven Model Context Protocol servers. Open-source, fully searchable.
claude mcp add --transport stdio el1fe-mcpdir npx -y mcpdir
How to use
MCP Hub (the MCPDIR project) is an open-source directory that indexes and validates MCP servers from multiple sources, providing a centralized place to discover tools that can be used by AI assistants to interact with external services, databases, and tools. The repository itself hosts the frontend and orchestration logic for syncing, validating, and displaying MCP servers, along with utilities for parsing MCP metadata and validating handshake compatibility. You can explore server cards, categories, and detailed server information, including recommended prompts, resources, and tools supported by each MCP server. The included tooling supports multi-source indexing, real-time GitHub integration, and a command palette to quickly find servers or actions. Users can run local development instances to preview the UI and test the synchronization/validation pipeline before contributing or submitting new MCP servers.
To use the available tools, clone the repository and start the development server. The project provides scripts for syncing servers from multiple sources, validating MCP handshakes inside sandboxed environments, and seeding category data. In practice, you would run the local dev server to browse, search, and inspect MCP servers, then leverage the components and APIs to integrate MCP data into your own tooling or automations. The UI exposes a search/command palette, server detail pages with metadata, and sections for categories and submission workflows, making it straightforward to experiment with MCP server discovery, validation, and documentation generation.
How to install
Prerequisites:
- Node.js 20+ and pnpm 9+
- Git
- PostgreSQL (or Neon) for full local setup (optional for frontend preview)
- Clone the repository
git clone https://github.com/eL1fe/mcpdir.git
cd mcpdir
- Install dependencies
pnpm install
- Set up environment (optional for full backend features)
cp .env.example .env.local
Configure your environment variables as needed (e.g., database URL, OAuth keys). For a minimal frontend preview, you can run the app with the provided scripts:
- Run the development server
pnpm dev
Open http://localhost:3000 to view the MCP Hub interface.
Notes:
- The project includes scripts for syncing and validating MCP servers; ensure Docker is available if you want to exercise the validation pipeline locally.
- For production-like validation, you may configure and run the validation commands described in the repository (e.g., pnpm validate).
Additional notes
Tips and common issues:
- Ensure Docker is running if you use Docker-based validation pipelines.
- When configuring environment variables, keep secrets out of source control and use a .env.local file for local development.
- The MCP Hub relies on multiple data sources (registry, npm, GitHub, Glama, PulseMCP); if a source is temporarily unavailable, you can still explore locally using the static UI and mocked data.
- If you encounter port conflicts, verify that the development server is not already running and that you are accessing the correct URL (http://localhost:3000 by default).
- The project supports a command palette and server search; use these features to quickly locate MCP servers, resources, and prompts.
- For contributors, the repo includes scripts for syncing and validating servers; consult the README for the latest commands and usage guidelines.
Related MCP Servers
fastapi_mcp
Expose your FastAPI endpoints as Model Context Protocol (MCP) tools, with Auth!
MakeMoneyWithAI
A list of open-source AI projects you can use to generate income easily.
gtm
An MCP server for Google Tag Manager. Connect it to your LLM, authenticate once, and start managing GTM through natural language.
akyn-sdk
Turn any data source into an MCP server in 5 minutes. Build AI-agents-ready knowledge bases.
extract-llms-docs
Extract documentation for AI agents from any site with llms.txt support. Features MCP server, REST API, batch processing, and multiple export formats.
mode-manager
MCP Memory Agent Server - A VS Code chatmode and instruction manager with library integration