s-hiraoku/claude-harnesses · Archived

review

Review code changes with a bug-first, risk-focused workflow before merge or release.

First seen Jun 13, 2026

Installation

$ npx skills add s-hiraoku/claude-harnesses --skill review

Stronger alternatives

This repository is archived — consider an actively maintained alternative.

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from s-hiraoku/claude-harnesses · top by installs.

npx skills add s-hiraoku/claude-harnesses

Browse all from s-hiraoku/claude-harnesses

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Repository health

License LICENSE
Default branch main
Open issues 1
Status Archived

Skill metadata

Parsed from SKILL.md frontmatter.

Declared agents claude-code

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 1,503 B
  • docs SUMMARY.md 98 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 1 installs

SKILL.md

Review

Use this workflow when reviewing a diff, pull request, branch, or local change set.

Workflow

  1. Identify the review scope: diff, files, commit range, pull request, or task goal.
  2. Inspect changed behavior before style or cleanup concerns.
  3. Prioritize bugs, regressions, security risks, data-loss risks, and missing verification.
  4. Check whether tests, docs, migrations, and release notes match the behavioral impact.
  5. Look for edge cases around empty input, permissions, concurrency, rollback, and failure paths.
  6. Keep findings specific, reproducible, and tied to exact files or lines when possible.
  7. Avoid broad refactor suggestions unless they block correctness, safety, or maintainability.
  8. If no issues are found, say so clearly and note remaining test gaps or residual risk.

When the review covers a large diff, consider delegating per-area passes to parallel subagents (for example a code-reviewer subagent per concern: correctness, security, performance, tests). Combine findings, deduplicate, and rank by severity.

Final Report

Lead with findings, ordered by severity.

For each finding, include:

  • severity or priority
  • file and line when available
  • what can fail
  • why it matters
  • a concrete fix direction

Then include:

  • open questions or assumptions
  • verification reviewed or still missing
  • brief summary of the reviewed change