AI Agent Index
ByHeather MacAvelia·Independently reviewed·Published Mar 27, 2026·Updated Nov 16, 2026
Independently verified against live vendor data on Oct 16, 2026.
Goose logo

by Agentic AI Foundation (created by Block)

Visit goose-docs.ai →

Open-source AI agent for code and general tasks, created by Block and now governed by the Agentic AI Foundation. Desktop app, CLI and API. Free under Apache 2.0, bring your own model.

How we scored it

Autonomy

4/5

Integrations

5/5

Pricing clarity

5/5

Evidence

4/5

Setup

4/5

The facts

Goose is an open-source, general-purpose AI agent that runs on the user machine as a desktop app for macOS, Linux and Windows, a command-line tool and an API. It writes and edits code, runs commands and tests, and also handles research, writing, automation and data analysis. It is built in Rust, licensed under Apache 2.0, and released roughly weekly. Block created Goose and donated it on April 7, 2026 to the Agentic AI Foundation (AAIF) at the Linux Foundation, which now develops it in the open alongside the Model Context Protocol and AGENTS.md. The repository has 56.7k stars and more than 450 contributors.

Goose itself is free and has no paid plan. Users bring their own model: it works with more than 15 providers, including Anthropic, OpenAI, Google, Ollama, OpenRouter, Azure and Amazon Bedrock, and it can also use an existing Claude, ChatGPT or Gemini subscription through the Agent Client Protocol (ACP). Model costs are paid to the chosen provider, and a built-in local inference option runs models fully on the machine with no API calls. Anonymous usage statistics can be switched off.

Goose connects to more than 70 extensions through the Model Context Protocol, such as GitHub, JetBrains, Playwright and Browserbase, and users can add skills or build their own extensions. Recipes capture a workflow as a portable YAML file to share with a team or run in CI. Goose also works as an ACP server, so editors such as Zed, JetBrains and VS Code can use it as their agent, and it can hand work to other ACP agents such as Claude Code and Codex. Organizations can build a custom distribution with preconfigured providers, extensions and branding, and an SDK exposes the provider layer to Rust, Python and Kotlin.

Developers who want a coding-focused terminal agent tied to one model vendor should consider Claude Code or OpenAI Codex. Developers who want another open-source, bring-your-own-model coding agent should look at Aider or OpenCode. Goose has no commercial support contract or compliance certifications of its own, so organizations that need either should evaluate a commercial product.

Current state Q4 2026: Goose is governed by AAIF with a public governance model of maintainers and core maintainers, the documentation lives at goose-docs.ai, and version 1.55.0 shipped on October 16, 2026.

Pricing

free

View pricing ↗

Segment

Business

Setup

moderate

Verified

Oct 16, 2026

Transparency

Public

Contract

—

Data training

Not Trained

Human in loop

Optional

Goose pricing

Goose is free to use. Checked against the vendor's own pages on Oct 16, 2026.

Goose features

Agentic codingTerminal agentAutonomousOpen sourceByok

Goose pros and limitations

Editorial assessment

Pros

  • ✓Free and open source under Apache 2.0 with foundation governance at the Linux Foundation, so teams can audit, fork and build their own distribution.
  • ✓Works with more than 15 model providers, existing Claude, ChatGPT or Gemini subscriptions through ACP, or fully local models, so no single model vendor is required.
  • ✓More than 70 MCP extensions plus recipes, skills and an ACP server mode let it run beyond coding and inside editors such as Zed, JetBrains and VS Code.

Limitations

  • –No commercial support or SLA: help comes from GitHub issues, Discord and community maintainers.
  • –No compliance certifications are published for the project, which matters for regulated teams that must use certified vendors.
  • –Model costs are separate and depend on the provider and usage, so teams have to set and watch their own provider spending limits.

Goose technical details

Deployment
desktopcliapi
Autonomous rateConfigurable: Goose runs multi-step tasks on its own and can ask for approval before tool use. The CLI and API support headless automation, and recipes can run in CI.
Integrations
GitHubJetBrainsZedVS CodeMCPPlaywrightBrowserbaseAnthropicOpenAIGoogle GeminiOllamaOpenRouterAzureAmazon Bedrock

Alternatives to Goose

Rating

4.3/ 5
★★★★★

Editorial score

How we score this →

Recognition

Transparent PricingListed 2026
Embed these on your site →

Built for

DevToolsOpen Source

Leave a review

Used Goose? Tell other buyers how it worked for you. Takes about two minutes.

Write a review

Compare

Related Content

Is this your product?

AI systems read this listing 39 times in the last 30 days. Last checked Oct 16, 2026. AI assistants repeat what this page says, so keep it current: Featured Listings are re-checked every 14 days.

Get Featured →Claim it free

AI Agent Price & Rating Tracker

Free, every two weeks. Price changes, new agent launches, acquisitions, and rating updates across the AI agent market. Verified against live vendor data, not vendor marketing.

No spam. Unsubscribe anytime. We never share your email. Read past issues