>goose | Your open source AI agent

# goose goose is an extensible, open source AI agent that enhances software development by automating coding tasks. It runs locally on your machine and connects to various tools and services through the Model Context Protocol (MCP). goose is available as both a desktop application and a command-line interface (CLI). ## Key Features - **Local-first**: Runs on your machine with full access to your development environment - **Extensible**: Add capabilities through MCP extensions (servers) - **Multi-provider**: Works with OpenAI, Anthropic, Google, Meta, Ollama, and more - **Cross-platform**: Available on macOS, Linux, and Windows - **Open source**: MIT licensed, community-driven development ## Core Concepts - **Extensions**: Plugins that give goose new capabilities via MCP servers (GitHub, Slack, databases, etc.) - **Recipes**: Reusable task templates that can be shared and automated - **Sessions**: Conversation contexts that maintain state across interactions - **Providers**: LLM backends that power goose (OpenAI, Anthropic, Ollama, etc.) - **Goosehints**: Project-specific instructions that customize goose behavior ## Documentation - Quickstart: /docs/quickstart - Installation: /docs/getting-started/installation - LLM Providers: /docs/getting-started/providers - Using Extensions: /docs/getting-started/using-extensions - CLI Commands: /docs/guides/goose-cli-commands - Recipes Guide: /docs/guides/recipes/ - Tutorials: /docs/tutorials/ - MCP Servers/Extensions: /docs/mcp/ ## Installation ### macOS/Linux/Windows (CLI) ```sh curl -fsSL https://github.com/aaif-goose/goose/releases/download/stable/download_cli.sh | bash ``` ### macOS (Homebrew) ```sh brew install --cask block-goose ``` Desktop applications are available for macOS, Linux, and Windows at /docs/getting-started/installation ## Common Use Cases - Automating repetitive coding tasks - Interacting with GitHub (PRs, issues, code review) - Database queries and management - File system operations - Web s