ReddRadar
Agent skill
requesting-code-review
Use when completing tasks, implementing major features, or before merging to verify work meets requirements. Dispatches three independent reviewers in parallel.
Install this agent skill to your Project
npx add-skill https://github.com/leegonzales/AISkills/tree/main/RequestingCodeReview/requesting-code-review
SKILL.md
Requesting Code Review
Dispatch three independent reviewers in parallel for multi-perspective code review.
Core principle: Three perspectives catch more than one. Dispatch all reviewers simultaneously.
Reviewers
| Reviewer | Skill/Tool | Strengths |
|---|---|---|
| Claude subagent | superpowers:code-reviewer Task tool |
Deep reasoning, plan compliance |
| Codex CLI | codex review --base <SHA> |
Line-level bugs, concise output |
| Gemini CLI | cat <<'EOF' | gemini |
Holistic view, large context (1M tokens) |
When to Request Review
Mandatory: After each task in subagent-driven development, after major features, before merge to main.
Optional: When stuck, before refactoring, after fixing complex bugs.
The Process
Step 1: Get git SHAs
BASE_SHA=$(git rev-parse HEAD~N) # or origin/main, or specific commit
HEAD_SHA=$(git rev-parse HEAD)
Step 2: Dispatch ALL THREE reviewers in parallel
All three run simultaneously. Use parallel tool calls.
2a. Claude subagent
Use Task tool with superpowers:code-reviewer type. Fill the template from the marketplace skill's code-reviewer.md with:
{WHAT_WAS_IMPLEMENTED}- What you built{PLAN_OR_REQUIREMENTS}- What it should do{BASE_SHA}/{HEAD_SHA}- Git range{DESCRIPTION}- Brief summary
2b. Codex CLI
codex review --base $BASE_SHA
This is Codex's built-in code review command. It automatically diffs from BASE_SHA to HEAD and provides line-level feedback.
2c. Gemini CLI
DIFF=$(git diff $BASE_SHA..$HEAD_SHA)
cat <<EOF | gemini
Code review request.
CONTEXT: [what was implemented and why]
REQUIREMENTS: [plan or spec reference]
DIFF:
$DIFF
Review for: code quality, architecture, error handling, security, testing gaps.
Categorize issues as Critical / Important / Minor with file:line references.
EOF
Step 3: Synthesize findings
After all three return, consolidate into a single report:
- Agreement — issues flagged by 2+ reviewers are high-confidence
- Unique findings — issues only one reviewer caught (still valid)
- Disagreements — where reviewers conflict, evaluate technically
- Severity — use the highest severity assigned by any reviewer
Step 4: Act on feedback
- Fix Critical issues immediately
- Fix Important issues before proceeding
- Note Minor issues for later
- Push back if reviewer is wrong (with reasoning)
Quick Reference
| Step | Action |
|---|---|
| Get SHAs | git rev-parse HEAD~N / git rev-parse HEAD |
| Claude | Task tool with superpowers:code-reviewer |
| Codex | codex review --base $BASE_SHA |
| Gemini | Pipe diff + context to gemini via heredoc |
| Synthesize | Consolidate all three, weight by agreement |
Integration with Workflows
Subagent-Driven Development: Review after EACH task. All three reviewers, every time.
Executing Plans: Review after each batch (3 tasks).
Ad-Hoc Development: Review before merge.
Red Flags
Never:
- Skip review because "it's simple"
- Use only one reviewer when all three are available
- Ignore Critical issues from any reviewer
- Proceed with unfixed Important issues
If a reviewer is unavailable (CLI not installed, API quota):
- Note which reviewer was skipped
- Proceed with remaining reviewers
- Do NOT skip review entirely
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
leegonzales/AISkills
context-continuity
High-fidelity context transfer protocol for moving conversations between AI agents. Preserves decision tempo, open loops, and critical context with graceful degradation. Use when the user says "transfer," "handoff," "continue this in another chat," or needs to work around context window limits. Produces structured artifacts (Minimal ~200 words, Full ~1000 words). DO NOT trigger on simple "summarize our conversation" requests—only when transfer intent is explicit.
leegonzales/AISkills
codex-peer-review
leegonzales/AISkills
silicon-doppelganger
Build psychometrically accurate personal proxy agents for the PAIRL Conductor system. Extracts personality, decision heuristics, and values into portable schemas that enable AI agents to negotiate, filter, and act on a principal's behalf.
leegonzales/AISkills
fabric-patterns
Run danielmiessler/fabric CLI patterns for content analysis, extraction, summarization, writing, security analysis, and more. Use when user asks to "use fabric," "run a pattern," "extract wisdom," "summarize with fabric," or when piping content through AI patterns would be more effective than inline processing. Triggers include "fabric," "pattern," "extract wisdom," "summarize this article," "analyze this threat report," or any reference to a specific fabric pattern name.
leegonzales/AISkills
moltbook-enclave
Secure, air-gapped interface for Moltbook (social network for AI agents). Isolates untrusted external content from your main agent's memory and context.
leegonzales/AISkills
sand-table
Design, scaffold, extract, and validate Sand Table simulations and event streams across domains. Meta skill that knows the protocol and all existing implementations.
Didn't find tool you were looking for?