How do I install the Mass Code Review skill?

Run npx skillmds@latest add maiconsouza89/mass-code-review in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Mass Code Review safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Mass Code Review?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Mass Code Review free to use?

Yes. Installing skills from SkillMD is free. This skill is licensed under CC-BY-4.

Who published Mass Code Review?

maiconsouza89 (@maiconsouza89) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Mass Code Review by maiconsouza89 — Agent Skill

Name: Mass Code Review
Rating: 4.2 (5 reviews)
Author: maiconsouza89

Mass Code Review

Reviews a pull request or a local diff for correctness bugs, missing tests and unsafe changes, and reports findings with file and line. Use when "review this PR", "review my changes" or "look for bugs in this diff". Do NOT use for writing the PR description (use mass-pr-description) or for a security-only audit of a skill (use mass-security-checklist).

maiconsouza89 Updated Sep 22, 2026

File contents

Code review

Find what would break, not what you would have written differently.

Steps

Get the diff. For a PR: gh pr diff <number> and gh pr view <number> --json title,body. For local work: git diff <base>...HEAD.
Read the linked issue or the PR summary first so the review judges the change against its own goal.
Walk the diff file by file and ask, for each hunk: what input makes this wrong, what happens on the error path, what test would have caught it.
Check the tests: does a new branch have an assertion that fails without the change? A test that only exercises the happy path is a finding.
Check the edges of the change: callers of a renamed or re-typed symbol, migrations against existing data, workflows that run the changed command.

Report

One finding per line, most severe first:

<severity> <path>:<line> - <what is wrong> - <input or state that triggers it>

Severities: bug (wrong result or crash), risk (unsafe under a plausible input), test (missing or weak coverage), nit (style, only if it hides a bug).

End with one line: No blocking findings. or Blocking: <count>. Do not post to GitHub unless the user asked; when asked, use gh pr review with the findings as the body.

Frequently asked questions