githubnextOfficial·104 Agent Skills Included

gh-aw

Write AI-powered GitHub Actions workflows in plain Markdown

Compiles Markdown files with YAML frontmatter into secure GitHub Actions workflows that run AI agents on repository events. Handles issue triage, pull request review, CI failure investigation, and scheduled reporting without writing complex YAML by hand. Sandboxes agent jobs as read-only by default and routes all writes through validated safe-output jobs with scoped permissions.

5,079|530

↗

npx skills add githubnext/gh-aw --all -g -y

Available:

Gives the AI agent global operating rules for this repository, including validation commands, lazy skill loading, and routing each task type to the correct local skill file.

All Skills in This Repository (104)

Pure Emerald Level Indicators

📦 In Repo

githubnext

playwright-cli

Automate browser interactions and generate Playwright test code through a command-line interface.

session-recovery

Detect and resume interrupted Copilot CLI sessions by querying the session_store SQLite database.

architectural-proposals

Write structured architectural proposals with problem statements, decisions, risks, and scope definitions.

pr-review-response

Replies to GitHub PR review comment threads after fixing issues via gh CLI and GraphQL.

distributed-mesh

Configures git-based state synchronization for squads distributed across multiple machines.

squad-commands

Presents a categorized interactive menu of Squad CLI operations for team and workflow management.

history-hygiene

Record final confirmed outcomes to history files instead of intermediate or reversed decisions.

ralph-two-pass-scan

Reduces GitHub API calls by filtering issue lists before selective hydration.

iterative-retrieval

Coordinates sub-agent task delegation with a 3-cycle retry protocol and coordinator validation.

fact-checking

Validate claims and deliverables using counter-hypothesis testing with structured review output.

docs-standards

Applies Microsoft Style Guide and Squad conventions to documentation writing.

personal-squad

Configure user-level AI agents that persist across projects via a global squad directory.

Official

Basic

Frequently Asked Questions

FAQPage Schema

How to install gh-aw?▼

Run `npx skills add githubnext/gh-aw --all -g -y` in your terminal to install all skills in this suite globally.

What does gh-aw do?▼

It is a GitHub CLI extension that turns Markdown files with YAML frontmatter into GitHub Actions workflows powered by AI agents. You describe the task in plain language and it compiles to a standard workflow file.

Which AI engines does gh-aw support?▼

It supports GitHub Copilot, Claude Code, OpenAI Codex, Google Gemini, and Pi. You select the engine in the workflow frontmatter and authenticate with the matching API key or token.

Is gh-aw safe to run on my repository?▼

Yes. Agent jobs are read-only and sandboxed by default, and any GitHub writes go through validated safe-output jobs with scoped permissions. You should still review permissions and generated files before deploying.

Do I need to know GitHub Actions YAML to use gh-aw?▼

No. You write the trigger, permissions, and task instructions in simple Markdown frontmatter, and `gh aw compile` generates the full Actions workflow for you.