code-review
Use when requesting code review, receiving code review feedback, before merging, or when handling reviewer suggestions - covers both giving and receiving review
How do I install this agent skill?
npx skills add https://github.com/eyadsibai/ltk --skill code-reviewIs this agent skill safe to install?
- Gen Agent Trust Hubpass
This skill provides a structured framework for AI agents to conduct and respond to code reviews. It emphasizes technical verification and clarification, including safety-oriented instructions to evaluate external feedback before implementation. No malicious patterns or security risks were identified.
- Socketpass
No alerts
- Snykpass
Risk: LOW · No issues
- Runlayerpass
1 file scanned · No issues
What does this agent skill do?
Code Review
Both sides of the code review process: requesting and receiving feedback.
When to Request Review
Mandatory
| Situation | Why |
|---|---|
| After completing major feature | Catch issues before they spread |
| Before merge to main | Gate quality |
| After each task in subagent workflow | Fix issues before compounding |
Valuable
| Situation | Why |
|---|---|
| When stuck | Fresh perspective |
| Before refactoring | Baseline check |
| After fixing complex bug | Verify fix doesn't introduce new issues |
Requesting Review
What to Provide
| Element | Purpose |
|---|---|
| What was implemented | Context for reviewer |
| Requirements/plan reference | What it should do |
| Base SHA | Starting point |
| Head SHA | Ending point |
| Brief summary | Quick orientation |
Acting on Feedback
| Severity | Action |
|---|---|
| Critical | Fix immediately |
| Important | Fix before proceeding |
| Minor | Note for later |
| Disagree | Push back with reasoning |
Receiving Review
Core Principle
Verify before implementing. Ask before assuming. Technical correctness over social comfort.
The Response Pattern
- Read - Complete feedback without reacting
- Understand - Restate requirement (or ask)
- Verify - Check against codebase reality
- Evaluate - Technically sound for THIS codebase?
- Respond - Technical acknowledgment or reasoned pushback
- Implement - One item at a time, test each
Handling Unclear Feedback
| Situation | Action |
|---|---|
| Some items unclear | STOP - ask before implementing any |
| Partially understood | Don't implement partial - items may be related |
| Scope unclear | Ask for clarification |
Key concept: Partial understanding leads to wrong implementation. Clarify everything first.
When to Push Back
| Situation | Push Back |
|---|---|
| Suggestion breaks existing functionality | Yes |
| Reviewer lacks full context | Yes |
| Unused feature (YAGNI violation) | Yes |
| Technically incorrect for this stack | Yes |
| Legacy/compatibility reasons exist | Yes |
| Conflicts with architectural decisions | Yes |
How to Push Back
| Do | Don't |
|---|---|
| Use technical reasoning | Be defensive |
| Ask specific questions | Argue emotionally |
| Reference working tests/code | Ignore valid feedback |
| Show evidence | Just say "no" |
Implementation Order
When fixing multiple items:
- Clarify anything unclear FIRST
- Then implement in order:
- Blocking issues (breaks, security)
- Simple fixes (typos, imports)
- Complex fixes (refactoring, logic)
- Test each fix individually
- Verify no regressions
Response Patterns
Forbidden (Performative)
| Don't Say | Why |
|---|---|
| "You're absolutely right!" | Performative, not technical |
| "Great point!" | Empty agreement |
| "Let me implement that now" | Before verification |
Correct Responses
| Situation | Response |
|---|---|
| Feedback is correct | "Fixed. [Brief description]" |
| Need clarification | "Need clarification on X before proceeding" |
| Disagree | "[Technical reasoning why current approach is better]" |
| Can't verify | "Can't verify without [X]. Should I investigate?" |
Key concept: Actions speak. Just fix it. The code shows you heard the feedback.
Red Flags
| Never Do | Why |
|---|---|
| Skip review because "it's simple" | Simple changes cause bugs too |
| Ignore Critical issues | They're critical for a reason |
| Proceed with unfixed Important issues | They'll compound |
| Argue with valid technical feedback | Ego over quality |
| Implement without understanding | Wrong fixes waste time |
Source-Specific Handling
| Source | Approach |
|---|---|
| User | Trusted - implement after understanding, skip to action |
| External reviewer | Evaluate technically before implementing |
| Automated tool | Verify relevance to your context |
How can the creator link this skill?
Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.
<a href="https://skillzs.dev/skills/eyadsibai/ltk/code-review">View code-review on skillZs</a>