Getting Started

Install Cerebro and connect it to your MCP client in under 2 minutes.

1Prerequisites

Python 3.10+
Check with python --version
pip
Usually included with Python. Check with pip --version

2Installation

Install Cerebro from PyPI with a single command:

terminal

pip install cerebro-ai

3Initialize

Run the init command to set up your storage directory and default configuration:

terminal

cerebro init

This creates the default storage at ~/.cerebro with the memory database, FAISS index, and config files.

4Configure Your MCP Client

Add Cerebro to your MCP client configuration. Here is the JSON snippet for Claude Desktop:

claude_desktop_config.json

{
  "mcpServers": {
    "cerebro": {
      "command": "cerebro",
      "args": ["serve"],
      "env": {
        "CEREBRO_STORAGE_PATH": "/path/to/your/storage"
      }
    }
  }
}

Config file locations:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Linux: ~/.config/Claude/claude_desktop_config.json

5Verify Installation

Confirm everything is working:

terminal

cerebro status

You should see a status report showing your storage path, memory count, FAISS index health, and available tools (49 total).

Next Steps

Explore the 49 MCP tools across 10 categories Customize storage, embeddings, and advanced settings