goose
What is goose?
TL;DRgoose is an open-source AI agent that runs locally through desktop, CLI, and API interfaces. It can handle coding, research, writing, automation, and data tasks while connecting to models, MCP extensions, subagents, and reusable workflow recipes.
- Open-source local AI agent
- Desktop app, CLI, and API
- MCP extensions and ACP support
Explore with AI
About
goose is an open-source AI agent that runs on your computer through desktop, CLI, and API interfaces. It can perform coding, research, writing, automation, data analysis, and other multi-step tasks.
goose connects to many model providers and supports extensions through the Model Context Protocol. Extensions can connect agents to databases, APIs, browsers, GitHub, cloud storage, and other tools.
The project includes reusable recipes, subagents, MCP Apps, and security controls such as permissions, sandboxing, and prompt-injection defenses. It supports macOS, Windows, and Linux and is released under Apache 2.0.
goose is a strong option for technical teams wanting an open, provider-flexible agent. Setup and governance requirements depend on the models, tools, permissions, and workflows connected to it.
Use Cases
Automate Coding Tasks
Use a local agent to write, edit, test, debug, and refactor software while keeping the workflow close to your development environment.
Run Local Agents
Build assistants that operate on your machine and connect to local files, tools, models, and services under your control.
Connect MCP Tools
Extend goose with MCP servers for databases, browsers, GitHub, cloud services, and other capabilities the agent can call.
Automate Research
Use the general-purpose agent for research, writing, data analysis, and repeatable workflows beyond traditional coding assistance.
Key Features
- Open-source local AI agent
- Desktop app, CLI, and API
- MCP extensions and ACP support
- Multi-provider model connectivity
We may earn commissions from links to support our work. Learn more.
Explore with AI
Pricing summary
goose is free and open source under Apache-2.0. Users may pay separately for external model providers, APIs, hosted infrastructure, or other connected services.
Pricing & Plans
Open Source
Free
- Apache 2.0
- Desktop app
- CLI access
- MCP extensions
Local Models
Free
- Local inference
- Offline workflows
- Local data processing
- No model API bill
Popular
Provider Usage
Custom
- Multiple providers
- API key support
- Subscription access
- Usage-based costs
Other pricing notes
- The goose software itself is free and open source.
- Model provider and external service charges are separate from the goose project.
Pricing last checked: August 2026Official pricing
Pros & Cons
Pros
- Open-source and vendor-neutral
- Runs locally across major desktop platforms
- Supports many model providers
- Large MCP extension ecosystem
- Useful beyond software development
Cons
- External model usage can still cost money
- Local setup requires technical familiarity
- Agent output needs human review
- Extension quality varies across providers
- Advanced workflows can require configuration
The Best goose Alternatives
The best goose alternatives are OpenClaw, Hermes Agent, and Claude Code.
Frequently Asked Questions
Reviews
Honest feedback from the FutureStack community.
0.00 ratings
No reviews yet. Be the first to share your experience.
Trending Tools on FutureStack
Discover the most popular and hand-picked AI tools across FutureStack
goose
AI Agents