Goose AI Agent

Description

Goose is a versatile open-source AI agent designed to run directly on your machine. It offers a native desktop application compatible with macOS, Linux, and Windows, a comprehensive command-line interface for terminal workflows, and an API for embedding Goose into various environments. Built in Rust, Goose ensures high performance and portability.

The agent is not limited to coding tasks; it can be used for research, writing, automation, data analysis, and more. Goose supports over 70 extensions, connecting to databases, APIs, browsers, GitHub, Google Drive, and more through the Model Context Protocol (MCP) open standard. Users can add skills or create their own extensions.

Goose works with more than 15 language model providers, including Anthropic, OpenAI, Google, Ollama, OpenRouter, Azure, and Bedrock. It allows the use of API keys or existing subscriptions like Claude, ChatGPT, or Gemini via the Agent Client Protocol (ACP).

The platform supports capturing workflows as portable YAML configurations, known as recipes, which can be shared, run in CI, and include instructions, extensions, parameters, and subrecipes. Goose also allows for the creation of interactive UIs within its desktop app through MCP Apps.

Security is a priority, with features like prompt injection detection, tool permission controls, sandbox mode, and an adversary reviewer to monitor for unsafe actions. Goose is part of the Agentic AI Foundation at the Linux Foundation, ensuring it remains vendor-neutral and community-governed.

Goose AI Agent's Core Features

• Open-source AI agent

• Desktop app for macOS, Linux, Windows

• Command-line interface

• API for embedding

• Supports 70+ extensions

• Integrates with 15+ LLM providers

• Portable YAML recipes

• Interactive UIs with MCP Apps

• Security features including prompt injection detection

• Part of the Agentic AI Foundation

How to use Goose AI Agent?

1. 1

   Download: Get the desktop app for your OS

2. 2

   Install: Follow installation instructions for setup

3. 3

   Configure: Set up extensions and integrations

4. 4

   Use: Execute tasks via desktop app, CLI, or API

5. 5

   Optimize: Share and refine workflows with recipes

Goose AI Agent's Use Cases

• Code Automation
• Data Analysis
• Research Assistance
• Writing Support
• Workflow Management

FAQ from Goose AI Agent

• What platforms does Goose support?

  Goose supports macOS, Linux, and Windows platforms.

• Can I use Goose for tasks other than coding?

  Yes, Goose can be used for research, writing, automation, and data analysis.

• What is the Model Context Protocol?

  MCP is an open standard for connecting AI agents to tools and data sources, supported by Goose.

• How does Goose ensure security?

  Goose includes prompt injection detection, tool permission controls, and sandbox mode for security.

• What is the Agent Client Protocol?

  ACP is a standard for communicating with coding agents, allowing Goose to connect with various providers.

• Can I create my own extensions for Goose?

  Yes, users can add skills or build their own extensions for Goose.

• What are MCP Apps?

  MCP Apps allow extensions to render interactive UIs directly inside the Goose desktop app.

• Is Goose part of any foundation?

  Yes, Goose is part of the Agentic AI Foundation at the Linux Foundation.