Skip to content

Skill Address Pr Review

Claude skill that fixes PR comments and CI failures

Install

$git clone https://github.com/mnapoli/skill-address-pr-review

Install command taken from the project's README or registry entry. Always review code before running it.

Address PR Review

An agent skill that reads pull request comments and CI failures and fixes them automatically.

Works in Claude Code, Codex, Cursor...

flowchart LR
    A[Open PR] --> B[GitHub code review]
    B --> C[`/address-pr-review`]
    C --> B
    B --> D[Merge PR]

Open a PR and invoke /address-pr-review in Claude Code (or $address-pr-review in Codex):

  • Fixes CI failures
  • Reads inline pull request comments (ignores resolved comments)
  • Makes code changes
  • Pushes and replies to each thread

Prerequisites

Installation

In Claude Code

Clone into your Claude Code skills directory:

git clone git@github.com:mnapoli/skill-address-pr-review.git ~/.claude/skills/address-pr-review

In other systems

Most other agent systems read from a ".agents/skills" directory:

git clone git@github.com:mnapoli/skill-address-pr-review.git ~/.agents/skills/address-pr-review

Usage

From a branch with an open PR:

/address-pr-review

Your agent will fetch unresolved threads and CI failures, fix the issues, and reply to reviewers.

Tips

  1. Run Claude Code's code review in GitHub Actions for automatic reviews on every push
  2. Run Codex code review as well
  3. Review Claude and Codex's comments and "mark as resolved" those that don't make sense
  4. Review the code yourself: open comments inline the PR diff
  5. Launch /address-pr-review locally so that all comments are addressed
  6. Repeat until the PR is ready to merge

Limitations

  • CI failure extraction is designed for GitHub Actions — other CI systems may not work

Similar plugs

Open-source AI job search agent and job finder: scan job boards, score each job 1-5 against your CV before you apply, tailor an ATS-friendly resume and cover letter, get…

Editorial diagram design for Claude Code, Codex, GitHub Copilot, Cursor, Factory Droid, and Pi. 44 diagram types. Self-contained HTML + SVG. No shadows. No Mermaid slop.

AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs

Secure, fast, efficient, battle-tested at Alibaba's scale. Hybrid architecture code review tool: deterministic pipelines + LLM Agent, precise line-level comments, built-in…

Persistent file-based planning for AI coding agents and long-running tasks. Crash-proof markdown plans, session recovery after /clear and compaction, per-turn re-injection…

Explore related