OrenAshkenazy

Gh Review Loop — AI skill for Claude Code

AI community

Fixes the AI reviewer findings on your PR, including the sibling instances the bot missed.

How to install Gh Review Loop

This entry records only its repository, not the path inside it, so there is no exact command to give. Open OrenAshkenazy/gh-review-loop and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Gh Review Loop does

Fixes the AI reviewer findings on your PR, including the sibling instances the bot missed. Runs your tests before it pushes. Hard capped at 3 rounds. Claude Code and Codex.

Alternatives in AI

  • Gtm Engineer Skills — Claude Code skill for improving website AEO (AI Engine Optimization) and GEO (Generative Engine Optimization) 761 ★
  • Openclaw Backup — One-click backup & restore for OpenClaw instances — workspace, credentials, skills, agent history 660 ★
  • Academic Writing Skills — AI-powered post-writing toolkit for academic papers — format validation, grammar/style polishing, de-AI editin 430 ★

README

gh-review-loop

**Your AI reviewer flagged two instances of a bug. There are five. This fixes all five, runs your tests, and stops after three rounds.**

[![CI](https://github.com/OrenAshkenazy/gh-review-loop/actions/workflows/ci.yml/badge.svg)](https://github.com/OrenAshkenazy/gh-review-loop/actions/workflows/ci.yml) [![Release](https://img.shields.io/github/v/release/OrenAshkenazy/gh-review-loop?sort=semver)](https://github.com/OrenAshkenazy/gh-review-loop/releases) [![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)

![Terminal demo of gh-review-loop clearing AI reviewer feedback on a PR](docs/gh-review-loop-demo.gif)

A skill for **Claude Code** and **Codex** that clears AI reviewer feedback on your PR without you leaving the terminal. It reads real GitHub review-thread state, fixes what's actionable, **sweeps the sibling instances the reviewer didn't flag**, gates every push on your repo's own tests, and asks for re-review, hard-capped so it can't spam the PR.

Works with whatever AI reviewer assistance your repo already has. MIT, no server, no account, no vendor plan.

Install

**Claude Code** — two slash commands:

/plugin marketplace add OrenAshkenazy/gh-review-loop
/plugin install gh-review-loop@gh-review-loop

**Codex:**

codex plugin marketplace add OrenAshkenazy/gh-review-loop
codex plugin add gh-review-loop@gh-review-loop

Then, from a repo with an open PR that a reviewer bot has commented on (the loop asks which bot the first time, and remembers):

Run the AI reviewer loop

That's the whole interface. [Other install paths](#other-install-paths) · [Prerequisites](#prerequisites)


What's actually implemented

Checkable claims, with the honest scope of each:

Capability Scope Where
Thread-state filtering reviewThreads GraphQL, isResolved / isOutdated, plus ADDRESSED_BY_REPLY for maintainer "wontfix" replies fetch_gemini_threads.py
Pattern clustering +