goose
an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Our Take
Block quietly built an open-source dev agent that actually does the work instead of just suggesting it, and the numbers don't lie — 35.9k stars on GitHub is the kind of organic traction that happens when a tool genuinely works. What's clever here is that Goose pairs with literally any LLM, runs as a desktop app or CLI, and its MCP server integration means you can plug it into your existing stack without locking yourself into someone's ecosystem. It's Apache-2.0 licensed, so there's no catch, and the multi-model configuration lets you optimize for speed or quality depending on what you're building — which is a real quality-of-life win for developers managing complex pipelines. If you're still manually triaging code suggestions while an agent could be scaffolding entire features for you, that's the kind of workflow friction that nobody should be tolerating in 2025.
A local, extensible, open source AI agent that automates engineering tasks. It can build entire projects from scratch, write and execute code, debug failures, orchestrate workflows, and interact with external APIs autonomously.
Key Facts
The people behind goose
Links
Want products like this in your inbox every morning?
Five products. Every morning. Written by someone who actually cares whether they're good or not. Free forever, unsubscribe whenever.