review-repair

Automate GitHub pull request repair workflows with in-thread replies and escalation.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/hack-ink/codexlab --skill review-repair
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-repair
Source: https://github.com/hack-ink/codexlab/tree/main/.codex/skills/review-repair
Command: npx skills add https://github.com/hack-ink/codexlab --skill review-repair

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

After a PR has received review feedback, managing unresolved threads and delivering repairs can be error-prone and time-consuming. This Skill automates the in-thread repair loop, handles replies, and escalates when external review cycles fail to converge.

Core Features & Use Cases

  • Triage unresolved review comments and group them into repair batches
  • Use the review-loop engine to apply repairs to the repaired diff and post in-thread replies
  • Resolve threads only when the fixes land and escalates to higher review channels when needed

Quick Start

Initiate a repair cycle for a PR by collecting unresolved threads and invoking the repaired-diff review loop.

Frequently Asked Questions about review-repair

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate GitHub pull request repair loops after review feedback?

Automate GitHub pull request repair loops by triaging unresolved review comments into batches, applying deterministic repairs via the review-loop engine, and posting in-thread replies. It coordinates the entire workflow from triage to resolution automatically.

What is the best way to triage and resolve unresolved GitHub PR review threads?

The best way to triage unresolved GitHub PR review threads is using an automated repair workflow that groups comments into repair batches, applies fixes, and conditionally resolves threads or escalates them when repairs stall.

How does automated thread escalation work when PR review cycles fail to converge?

Automated thread escalation triggers when repair cycles stall after applying deterministic repair batches. The workflow detects failed convergence and escalates unresolved threads to higher review channels automatically.

Do I need a stable head_sha verification to apply repaired diffs in GitHub PR automation?

Yes, stable head_sha verification is required for each repaired diff in GitHub PR automation. This ensures the repair batch applies cleanly to the correct commit state before posting in-thread replies.

Can I integrate review-loop automation with existing GitHub PR data for in-thread replies?

Yes, review-loop automation integrates directly with GitHub PR data to collect unresolved threads, apply deterministic repair batches, and post in-thread replies with conditional thread resolution based on repair outcomes.

What are the limitations of automating PR repair workflows when external review feedback stalls?

Automating PR repair workflows cannot force convergence when external review feedback continuously stalls. The system handles this by escalating unresolved threads to higher review channels rather than indefinitely retrying repairs.