Tools/aaif-goose/goose

goose

an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM

56.7k ★+291/wkestablishedRustApache License 2.0 →new release v1.29.1

The Lens

By Erik Loyd, SaaS CEO and former COO/CFO of an AWS Premier Partner.

Updated Oct 2026

goose is an AI agent that runs on your machine and does the work instead of suggesting it: it edits files, runs commands and calls tools, for code or for research and data jobs. Block built it, then in April 2026 donated it to the Linux Foundation's Agentic AI Foundation. Apache 2.0, free, as a desktop app, CLI or API.

Nothing to host. It plugs into 70+ MCP extensions and 15+ model providers. Two cheap paths are new: built-in local models with no Ollama and no API key, or your existing Claude Code or ChatGPT Plus/Pro subscription over ACP instead of per-token billing. Local models are slower and weaker at tool calling.

There's no paid tier; the model is your only bill. Solo: free. Small teams: free, share recipes (reusable YAML workflows). Larger teams: free, and custom distributions let you ship a preconfigured internal build.

The catch: autonomous mode is the default, so out of the box goose can modify and delete files without asking. Switch to Smart Approval first. In July, a high-severity advisory showed goose review in a malicious repo could run an attacker's command before any prompt (fixed in 1.44.0).

Free vs Self-Hosted vs Paid

fully free

Free: Everything. Apache 2.0 desktop app (macOS, Linux, Windows), CLI and API, with no paid edition and no hosted service. Block donated the project to the Agentic AI Foundation at the Linux Foundation in April 2026; the repo moved from block/goose to aaif-goose/goose.

Self-hosted: It runs locally, so there is nothing to host. Pick a model source: built-in local inference (llama.cpp in-process, downloads GGUF models from Hugging Face, no API key, free after the download), Ollama, API keys for 15+ providers (Anthropic, OpenAI, Google, OpenRouter, Azure, Bedrock and more), or ACP providers that reuse an existing Claude Code, ChatGPT Plus/Pro, Amp or Pi subscription with no per-token API costs. The goose team's own notes on local models: slower, most work with 4K to 8K tokens of context, and tool calling is the gap.

Paid: Nothing to buy from goose. Your spend is the model: per-token API pricing, or the flat subscription you already pay for through ACP. ACP providers currently can't resume or fork sessions.

Free agent. The model is the bill, and you can now cap it by reusing a Claude Code or ChatGPT Plus/Pro subscription over ACP, or zero it with built-in local models.

What to do by team size

Solo
free; reuse your Claude Code or ChatGPT subscription over ACP
Small team
free; share recipes and set Smart Approval as the default mode
Medium team
free; standardize providers and permission modes across machines
Large team
free; ship a preconfigured custom distribution internally
Self-hosting ops:trivial

Get tools like this every Wednesday

One featured tool, three on the radar. No fluff.

Similar Tools

Score
100/100 · A+
Adoption30/30
Maintenance25/25
Community20/20
License15/15
Analysis10/10

A low score is not a verdict on quality. Young and niche tools start low by design. How we calculate scores

Trust Signals

High adoption: 33,422 starsActive community: 3,106 forksCommunity discussions enabledOrganization account (107 public repos)Notable author: 1,211 followersBacked by: Block Inc (fka Square)

License: Apache License 2.0

Use freely. Patent grant included.

Commercial use: ✓ Yes

About

Owner
Block Open Source (Organization)
Backed by
Block Inc (fka Square)
Stars
56,675
Forks
6,593
Reddit
trending

Explore Further

More tools in the directory

Featured in The Open Source Drop #1