Run your first Cognee MCP server to see AI memory integration in action.

​
Prerequisites

  • Docker installed and running
  • OpenAI API key

​
Quick Setup

1

Set your API key

export LLM_API_KEY=your_api_key_here
2

Create .env file

echo "LLM_API_KEY=your_api_key_here" > .env
3

Launch the server

docker run -e TRANSPORT_MODE=http --env-file ./.env -p 8000:8000 --rm -it cognee/cognee-mcp:main
4

Test the server

curl http://localhost:8000/health
The server runs on port 8000. By default, the container removes all data when stopped.

​
Connect to AI Tools

Cognee MCP is designed to work with AI development tools. Choose your preferred client:

Cursor

AI-powered code editor with built-in MCP support

Claude Code

Claude’s coding assistant with MCP integration

Continue

Open-source AI coding assistant

Cline

Command-line AI assistant

Roo Code

AI-powered development environment

​
Next Steps

Tools Reference

Complete reference for all available MCP tools and operations

Local Setup

Deploy from source with full control and latest features