Welcome to Agent MCP

What is Agent MCP?

Agent MCP is an open source project directory that focuses solely on AI agents and MCP orchestration, distinguishing itself from platforms like Hugging Face which focus on LLMs and datasets. It is available at www.agentmcp.ai.

How to use Agent MCP?

Browse and search the directory directly at www.agentmcp.ai. To edit the code locally, clone the repository, run npm i to install dependencies, and start the development server with npm run dev. A Firebase project with Google Authentication and Firestore must be configured for user authentication and search history features.

Key features of Agent MCP

• Browse and search AI Agent and MCP repositories
• Bulk import repositories from GitHub
• User authentication with Google
• Save and re-import previous search history
• Supports CursorAI, Windsurf AI, and Trey AI

Use cases of Agent MCP

• Discover and explore AI agent projects and MCP orchestration tools
• Maintain a personalized search history for repeated reference
• Bulk import GitHub repositories for centralized organization
• Compare and find agent-focused projects distinct from LLM datasets

FAQ from Agent MCP

How does Agent MCP differ from Hugging Face?

Agent MCP focuses specifically on AI agents and MCP orchestration, whereas Hugging Face primarily covers LLMs and datasets.

What technologies are used to build Agent MCP?

The project is built with Vite, TypeScript, React, shadcn-ui, and Tailwind CSS.

How do I set up Firebase authentication?

Create a Firebase project, enable Google Authentication, create a Firestore database, obtain your Firebase configuration, copy the .env.example file to .env.local, and fill in the configuration values.

What are the requirements for local development?

You need Node.js and npm installed. Clone the repo, run npm i, then npm run dev to start the development server.