gh-aw vs goon

AI workflow automation compared

gh-aw adds AI reasoning to repository automation by compiling Markdown workflow definitions into GitHub Actions. goon runs as a self-hosted worker for software tasks and custom workflows, with local state and approval options.

gh-awgoon
LanguageGoGo
LicenseMITMIT
Stars5.4k3
Forks5960
Last analyzedOct 23, 2026Oct 17, 2026

Key differences

  • gh-aw focuses on interpreting repository issues, pull requests, and CI results within GitHub Actions; goon also supports scheduled workflows and Jira issue polling.
  • gh-aw defines workflows in Markdown with YAML frontmatter and compiles them into GitHub Actions files; goon runs as a daemon or CLI and supports custom JSON workflows.
  • gh-aw uses read-only, sandboxed agent jobs by default and routes configured writes through safe outputs; goon offers approval gates and configurable execution modes.
  • gh-aw supports GitHub Copilot, Claude Code, OpenAI Codex, Google Gemini, and Pi; goon supports OpenAI, Anthropic, Gemini, and Ollama with per-role model routing.
  • Both projects are written in Go and use the MIT license; the provided facts list gh-aw at 5.3k stars and 572 forks, and goon at 3 stars and 0 forks.

Choose gh-aw if you…

  • need AI reasoning within an existing GitHub Actions setup.
  • want Markdown-defined workflows compiled into standard GitHub Actions files.
  • prefer read-only, sandboxed agent jobs by default with scoped write handling.
Read the gh-aw analysis →

Choose goon if you…

  • want a self-hosted daemon for ongoing or scheduled workflows.
  • need approval options through a web UI, Telegram, or terminal.
  • want local project notes and run history, with integrations beyond GitHub.
Read the goon analysis →

Compare the numbers side by side, or add more projects

This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.