Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.cognee.ai/llms.txt

Use this file to discover all available pages before exploring further.

Goose Integration

Goose is an open-source AI coding assistant by Block with built-in MCP support. It runs as a CLI or desktop app and connects to MCP servers through its extensions system.

Prerequisites

Setup Steps

1

Open the Goose configuration file

Goose stores its configuration at ~/.config/goose/config.yaml. Create the file if it does not exist:
mkdir -p ~/.config/goose
touch ~/.config/goose/config.yaml
2

Add the Cognee extension

Choose the configuration that matches how you started the Cognee MCP server:
Use this if you started the server with Docker:
extensions:
  cognee:
    name: cognee
    type: sse
    uri: http://localhost:8000/mcp
    enabled: true
3

Start Goose

goose session
Goose will load the Cognee extension on startup. You can verify it is active by running /tools inside the session — Cognee tools should appear in the list.
4

Use Cognee Tools

Goose will invoke Cognee tools automatically when relevant. You can also ask explicitly:
  • “Remember this file in Cognee”
  • “Recall authentication patterns from Cognee”
  • “Improve the current Cognee dataset”

Where to Use This Configuration

The ~/.config/goose/config.yaml file is the global Goose configuration. Extensions defined here are available in all Goose sessions.

Need Help?

Join Our Community

Get support and connect with other developers using Cognee MCP.