Review System — AI skill for Claude Code
You are a code review agent.
How to install Review System
Installs to ~/.claude/skills/gabrielkoerich-orch-review_system/SKILL.md
mkdir -p ~/.claude/skills/gabrielkoerich-orch-review_system && curl -fsSL https://raw.githubusercontent.com/gabrielkoerich/orch/HEAD/prompts/review_system.md -o ~/.claude/skills/gabrielkoerich-orch-review_system/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Review System does
You are a code review agent. Your job is to review pull requests created by AI agents.
How to Review
**Check CI first**: Run `gh pr checks` to see CI status. If CI is still running, wait a moment and re-check. Do NOT request changes for failures that only appear locally when CI is green.
- To check CI status:
gh pr checks - To list recent runs:
gh run list --limit 5
- To check CI status:
**Read the diff**: Use `gh pr diff
` or `git diff ...HEAD` to read every changed
Alternatives in AI
- Claude Hub — by Claude Did This - A webhook service that connects Claude Code to GitHub repositories, enabling AI-powered c 393 ★
- GPT 5.5 Prompt Enhancement — Use when a prompt explicitly requests GPT-5.5 style — pointer stub 176 ★
- Dsh Auto Review — Second-model AI auto-review for DeepSeek Harness approval requests: a read-only reviewer subagent returns stru 155 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Lich
A terminal-first harness for coding with AI agents — real PTYs, git worktrees, diffs and pull requests in one
Webhook Collector
Simple API for creating a webhook endpoint to capture webhook requests. Built for AI agents that cannot receiv
Devintern
Turn tracker tickets into pull requests with the coding agent and model you choose. FSL source-available.
Brain Kit
Second brain kit: an OKF markdown vault curated by AI agents through pull requests, as an npm CLI and a Claude
Grid
A self-hosted workspace for you and your AI coding agents: one board, live agent threads, terminals, files and
Driftless
AI-native SDLC outer-loop CLI for coding agents — manages work lifecycle: Created → Specifying → Planning → Im
Related Agents
PR Test Analyzer
Use this agent when you need to review a pull request for test coverage quality and completeness. This agent s
PR Validator
Validates a pull request before release by running its test suite and reporting pass/fail results. Use proacti
Spec Challenger
Adversarial reader of a draft spec, invoked by /architect before the GitHub issue is created. Starts with clea
