goose
An open-source, extensible AI agent that runs locally for code, workflows, and automation with support for 15+ LLM providers.
What is goose?
goose is an open-source AI agent that runs locally on your machine, designed to assist with code, research, writing, automation, and data analysis. It provides a desktop app, CLI, and API, allowing users to interact with multiple LLM providers without relying on cloud-based services. The project solves the need for a flexible, extensible AI tool that can be integrated into various workflows while maintaining data privacy and control.
Developers, researchers, and technical users who need a local AI agent for code assistance, workflow automation, and general task automation across macOS, Linux, and Windows platforms.
Developers choose goose for its local execution, support for 15+ LLM providers, extensibility via the Model Context Protocol, and the ability to create custom distributions. Its Rust-based architecture ensures performance and portability, making it a versatile alternative to cloud-only AI tools.
Overview
an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Use Cases
Best For
- Running AI agents locally for privacy-sensitive tasks
- Integrating AI into terminal-based development workflows
- Building custom AI-powered desktop applications
- Automating research, writing, and data analysis tasks
- Connecting to multiple LLM providers through a single interface
- Creating branded AI tool distributions with preconfigured extensions
Not Ideal For
- Users seeking a no-configuration, immediately usable AI assistant without setting up API keys
- Teams requiring real-time cloud-based collaboration features for shared AI workflows
- Projects running on resource-constrained devices like low-end laptops or embedded systems
- Organizations needing out-of-the-box integrations with enterprise SaaS platforms like Salesforce or Jira
Pros & Cons
Pros
Provides native desktop applications for macOS, Linux, and Windows, ensuring a consistent and user-friendly interface across all major operating systems as highlighted in the README.
Works with 15+ providers including Anthropic, OpenAI, and Google via API keys or existing subscriptions, offering flexibility in AI model choice without vendor lock-in.
Connects to 70+ extensions using the Model Context Protocol, enabling enhanced functionality through an open standard that avoids proprietary ecosystems.
Allows building personalized versions with preconfigured providers, extensions, and branding, ideal for organizations needing tailored deployments, as documented in CUSTOM_DISTROS.md.
Cons
Requires manual setup of API keys for LLM providers, which can be tedious and a barrier for users expecting plug-and-play functionality compared to cloud-based alternatives.
Running AI models locally can be computationally intensive, potentially limiting performance on older or low-powered hardware, despite Rust's efficiency.
The recent move to the Agentic AI Foundation has left some links and references outdated, as noted in the README, which may hinder onboarding during the transition.
Frequently Asked Questions
Related Projects
just🤖 Just a command runner
jjA Git-compatible VCS that is both simple and powerful
misedev tools, env vars, task runner
difftastica structural diff that understands syntax 🟥🟩
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.