#4343rd of 87 in Coding agents and assistants
gh-aw
GitHub CLI extension that compiles Markdown agent workflows into GitHub Actions
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 5.4k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- Go
3,975 on 25 Mar 20265,399 on 14 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
gh-aw is a GitHub CLI extension. You write repository automation as Markdown with YAML frontmatter, and `gh aw compile` turns it into a `.lock.yml` GitHub Actions workflow that runs an AI agent. Built-in engines are GitHub Copilot, Claude Code, OpenAI Codex, Google Gemini and Pi. Agent jobs are read-only and sandboxed by default, and writes go through validated `safe-outputs` jobs.
Who it is for: Teams on GitHub automating triage, review and CI investigation with AI agents
Strengths
- Compiles to standard Actions workflows, so it runs alongside existing CI/CD
- Supports Copilot, Claude Code, Codex, Gemini and Pi engines
- Agent jobs are read-only by default; writes are applied by separate scoped safe-outputs jobs
- Targets tasks like issue triage, PR review, CI failure investigation and doc maintenance
Weaknesses
- Runs only on GitHub Actions and GitHub repositories
- README says use is at your own risk and needs careful human supervision
- Releases >= 0.83.3, < 0.85.4 were retired over a security vulnerability
- Permissions, tools and network access are configurable, so authors must review them
What it needs
- no GPU
- Docker
- Needs GitHub CLI, GitHub Actions
- Models: GitHub Copilot, Claude Code, OpenAI Codex, Google Gemini, Pi
Also in Coding agents and assistants
See all 87| Rank | Project | Score |
|---|---|---|
| 1 | claw-codeRust CLI coding agent harness using Anthropic and OpenAI-compatible APIs | 83 out of 100 |
| 2 | ArchonYAML workflow engine that runs AI coding agents through fixed dev steps | 82 out of 100 |
| 3 | oh-my-openagentTerminal coding agent with multi-model orchestration and git-backed memory | 80 out of 100 |
| #4 | Open SWELangChain coding agent that plans, implements and reviews pull requests | 80 out of 100 |
| #5 | opencodeTerminal coding agent with build and plan modes, plus a desktop app | 79 out of 100 |