requesting-code-review · bobmatnyc — Claude Code skill: subagent code review
This skill makes the agent dispatch a code-reviewer subagent and check the work against the plan or requirements before moving on.
npx skills add bobmatnyc/claude-mpm-skills --skill requesting-code-review -a claude-codeWhat requesting-code-review does
The skill spells out when a review should happen: after every task in subagent-driven development, after a major feature, before merging to main. It also covers the optional cases — being stuck, refactoring, or a tricky bug you just fixed. The stated principle is to review early and often.
Mechanically it is short. You grab the base and head SHAs with git rev-parse, fill in the template from references/code-reviewer-template.md, and hand it to a code-reviewer subagent through the Task tool. Five placeholders are required: {WHAT_WAS_IMPLEMENTED}, {PLAN_OR_REQUIREMENTS}, {BASE_SHA}, {HEAD_SHA} and {DESCRIPTION}. Feedback lands in three buckets — Critical gets fixed immediately and blocks progress, Important waits until the next major task, Minor can be noted for later. If the reviewer is wrong, the skill tells you to push back with technical reasoning and references to the plan.
One thing to know before installing: the frontmatter sets user-invocable: false and disable-model-invocation: true, so there is no slash command here. The agent picks the skill up on its own. Two reference files ship with it, a subagent template and a set of review examples.
What it helps with
- check a new feature against requirements before merge
- catch issues after each task before they compound
- get a second opinion when stuck on a problem
- take a baseline before refactoring
- review the fix for a complex bug
Example prompts for the agent
- dispatch the code-reviewer subagent and check my changes against the plan
- review the commits between origin/main and HEAD before I merge
- fill in the code-reviewer template and run a review of this task
How to install requesting-code-review
1Skills CLI (Vercel)
npx skills add bobmatnyc/claude-mpm-skills --skill requesting-code-review -a claude-code2Manually: put the files in place
~/.claude/skills/requesting-code-review/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/requesting-code-review/
1Skills CLI (Vercel)
npx skills add bobmatnyc/claude-mpm-skills --skill requesting-code-review -a codex2Manually: put the files in place
.agents/skills/requesting-code-review/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add bobmatnyc/claude-mpm-skills --skill requesting-code-review -a cursor2Manually: put the files in place
.cursor/skills/requesting-code-review/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install bobmatnyc/claude-mpm-skills requesting-code-review2Manually: put the files in place
.github/skills/requesting-code-review/SKILL.md1Skills CLI (Vercel)
npx skills add bobmatnyc/claude-mpm-skills --skill requesting-code-review -a gemini-cli2Platform CLI
gemini skills install https://github.com/bobmatnyc/claude-mpm-skills --path universal/collaboration/requesting-code-reviewManually: put the files in place
.devin/skills/requesting-code-review/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add bobmatnyc/claude-mpm-skills --skill requesting-code-review -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about requesting-code-review
How do I install requesting-code-review in Claude?
Run npx skills add bobmatnyc/claude-mpm-skills --skill requesting-code-review -a claude-code. The skill lands at ~/.claude/skills/requesting-code-review/SKILL.md.
What does requesting-code-review do?
It dispatches a code-reviewer subagent that compares the implementation to the plan or requirements. Findings are sorted into Critical, Important and Minor, each with its own follow-up rule.
Why can't I invoke this skill manually?
The frontmatter sets user-invocable: false and disable-model-invocation: true. Reviews are triggered by the agent during its work, not by a slash command.
Source text
SKILL.md · MIT
---
name: requesting-code-review
description: Dispatch code-reviewer subagent to review implementation against plan or requirements before proceeding
user-invocable: false
disable-model-invocation: true
when_to_use: when completing tasks, implementing major features, or before merging, to verify work meets requirements
version: 1.2.0
progressive_disclosure:
level: 1
references:
- path: references/code-reviewer-template.md
title: Code Reviewer Template
description: Complete subagent template with placeholders and review checklist
- path: references/review-examples.md
title: Review Examples & Workflows
description: Good vs bad reviews, severity guidelines, complete workflow examples
---
# Requesting Code Review
Dispatch code-reviewer subagent to catch issues before they cascade.
**Core principle:** Review early, review often.
## When to Request Review
**Mandatory:**
- After each task in subagent-driven development
- After completing major feature
- Before merge to main
**Optional but valuable:**
- When stuck (fresh perspective)
- Before refactoring (baseline check)
- After fixing complex bug
## Quick Start
**1. Get git SHAs:**
```bash
BASE_SHA=$(git rev-parse HEAD~1) # or origin/main
HEAD_SHA=$(git rev-parse HEAD)
```
**2. Dispatch code-reviewer subagent:**
Use Task tool with code-reviewer type, fill template at [Code Reviewer Template](references/code-reviewer-template.md)
**Required placeholders:**
- `{WHAT_WAS_IMPLEMENTED}` - What you just built
- `{PLAN_OR_REQUIREMENTS}` - What it should do
- `{BASE_SHA}` - Starting commit
- `{HEAD_SHA}` - Ending commit
- `{DESCRIPTION}` - Brief summary
**3. Act on feedback:**
| Severity | Action |
|----------|--------|
| **Critical** | Fix immediately, don't proceed |
| **Important** | Fix before next major task |
| **Minor** | Note for later, can proceed |
See [severity guidelines](references/review-examples.md#severity-guidelines) for details.
## Integration with Workflows
**Subagent-Driven Development:**
- Review after EACH task
- Catch issues before they compound
- Fix before moving to next task
**Executing Plans:**
- Review after each batch (3 tasks)
- Get feedback, apply, continue
**Ad-Hoc Development:**
- Review before merge
- Review when stuck
## Pushing Back on Reviews
**If reviewer wrong:**
- Push back with technical reasoning
- Show code/tests that prove it works
- Reference plan requirements
- Request clarification
See [pushing back examples](references/review-examples.md#pushing-back-on-reviews)
## Common Mistakes
**Never:**
- Skip review because "it's simple"
- Ignore Critical issues
- Proceed with unfixed Important issues
- Argue without technical justification
**Always:**
- Provide full context in review request
- Fix Critical issues immediately
- Document disagreements with technical justification
- Re-review after fixing Critical issues
## Examples
**Need examples?** See [Review Examples & Workflows](references/review-examples.md) for:
- Complete review output examples
- Good vs bad review requests
- Review workflows for different scenarios
- How to act on different severity levels
- When and how to push back
**Need template?** See [Code Reviewer Template](references/code-reviewer-template.md) for the complete subagent dispatch template.
© bobmatnyc · MIT · bobmatnyc/claude-mpm-skills
Similar skills
Everything for Code review and refactoring →requesting-code-review by other authors comes first.