Back to Workflow & Productivity
requesting-code-review
code reviewquality assurancegit workflowcollaborationdevelopment processAI agentsoftware engineeringcoding standards
Install this skill
npx skills add obra/superpowersWorks across Claude Code, Cursor, Codex, Copilot & Antigravity
The requesting-code-review skill acts as a quality assurance gatekeeper within the superpowers subagent ecosystem. It orchestrates the code-reviewer subagent to inspect diffs against a project's original intent, ensuring that implementations match requirements before moving further. By supplying git SHAs and specific task metadata, this skill identifies critical defects, architectural gaps, and adherence to established plans. It forces a pause in the development loop, requiring developers to address feedback before proceeding to subsequent tasks or merging into the main branch. This approach minimizes technical debt accumulation by validating every iteration of a subagent-driven workflow. It transforms code auditing from an afterthought into an essential, repeatable step, ensuring that each discrete piece of work is verified, stable, and ready for integration.
When to Use This Skill
- β’Verifying completion of individual subagent tasks
- β’Performing a baseline check before initiating a major refactor
- β’Conducting final quality assurance prior to a merge request
- β’Troubleshooting when development momentum stalls or results seem incorrect
How to Invoke This Skill
Example prompts that trigger this skill in Claude Code, Cursor, or Antigravity:
- βReview my last task against the current plan
- βInitiate code review for the latest changes
- βCheck my progress before moving to the next task
- βRun code-reviewer on the latest commit
- βValidate the current implementation before I merge
Pro Tips
- π‘Always provide clear and concise context in the `{DESCRIPTION}` placeholder, explaining the 'why' behind the changes, not just the 'what'.
- π‘Utilize the `BASE_SHA` and `HEAD_SHA` effectively to define a precise review scope, especially when making incremental changes to a larger feature.
- π‘Don't hesitate to initiate reviews even for small, self-contained changes; embracing the 'review early, review often' principle minimizes technical debt over time.
What this skill does
- β’Validates implementation against project plans
- β’Identifies critical and minor bugs via diff analysis
- β’Forces mandatory checkpointing between development tasks
- β’Standardizes feedback cycles through template-driven reporting
- β’Enables evidence-based technical pushback against reviewer suggestions
When not to use it
- βMinor configuration adjustments that do not impact application logic
- βInitial drafting where the codebase is in a transient, non-functional state
Example workflow
- Finalize the current code task within the active branch
- Extract the starting and ending commit SHAs using git rev-parse
- Trigger the code-reviewer subagent with the defined template
- Review the returned assessment for critical or important issues
- Apply required fixes and verify with a follow-up review
Prerequisites
- βA clean git working directory
- βA defined task plan or requirements document
- βThe superpower:code-reviewer subagent agent installed
Pitfalls & limitations
- !Skipping reviews for simple tasks leads to cascading logic errors
- !Ignoring feedback categorized as 'Important' complicates later integration
- !Providing incorrect commit SHAs results in inaccurate diff analysis
FAQ
Do I have to fix minor issues immediately?
No, minor issues can be noted for future maintenance, but critical and important issues must be resolved before proceeding.
What if I disagree with the subagent feedback?
You should push back by providing technical reasoning, tests, or code examples that clarify why your implementation is correct.
How often should I request a review?
Mandatory reviews should occur after every subagent task, major feature completion, and before final code merges.
Can I request a review if I am stuck?
Yes, requesting a review can provide a fresh perspective on a complex bug or a stalled implementation.
How it compares
Unlike a generic prompt asking 'is this code okay,' this skill enforces a structured, template-based audit that treats the subagent as an official quality gate rather than a conversational assistant.
π Full skill instructions β original source: obra/superpowers
# Requesting Code Review
Dispatch superpowers: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
## How to Request
**1. Get git SHAs:**
**2. Dispatch code-reviewer subagent:**
Use Task tool with superpowers:code-reviewer type, fill template at
**Placeholders:**
-
-
-
-
-
**3. 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)
## Example
## 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
## Red Flags
**Never:**
- Skip review because "it's simple"
- Ignore Critical issues
- Proceed with unfixed Important issues
- Argue with valid technical feedback
**If reviewer wrong:**
- Push back with technical reasoning
- Show code/tests that prove it works
- Request clarification
See template at: requesting-code-review/code-reviewer.md
Dispatch superpowers: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
## How to Request
**1. Get git SHAs:**
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 superpowers:code-reviewer type, fill template at
code-reviewer.md**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:**
- Fix Critical issues immediately
- Fix Important issues before proceeding
- Note Minor issues for later
- Push back if reviewer is wrong (with reasoning)
## Example
[Just completed Task 2: Add verification function]
You: Let me request code review before proceeding.
BASE_SHA=$(git log --oneline | grep "Task 1" | head -1 | awk '{print $1}')
HEAD_SHA=$(git rev-parse HEAD)
[Dispatch superpowers:code-reviewer subagent]
WHAT_WAS_IMPLEMENTED: Verification and repair functions for conversation index
PLAN_OR_REQUIREMENTS: Task 2 from docs/plans/deployment-plan.md
BASE_SHA: a7981ec
HEAD_SHA: 3df7661
DESCRIPTION: Added verifyIndex() and repairIndex() with 4 issue types
[Subagent returns]:
Strengths: Clean architecture, real tests
Issues:
Important: Missing progress indicators
Minor: Magic number (100) for reporting interval
Assessment: Ready to proceed
You: [Fix progress indicators]
[Continue to Task 3]## 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
## Red Flags
**Never:**
- Skip review because "it's simple"
- Ignore Critical issues
- Proceed with unfixed Important issues
- Argue with valid technical feedback
**If reviewer wrong:**
- Push back with technical reasoning
- Show code/tests that prove it works
- Request clarification
See template at: requesting-code-review/code-reviewer.md
How to Use This Skill Unit
Option A: Project-Specific (Recommended)
- Click "Download" above
- In your project, create the directory:
.agent/skills/requesting-code-review/ - Save the file as
SKILL.md - The agent will automatically discover the skill based on its description.
Option B: Global Installation (All Agents)
Save the file to these locations to make it available across all projects:
- Claude Code:
~/.claude/skills/obra/superpowers/requesting-code-review/SKILL.md - Cursor:
~/.cursor/skills/obra/superpowers/requesting-code-review/SKILL.md - Antigravity:
~/.gemini/antigravity/skills/obra/superpowers/requesting-code-review/SKILL.md
π Install with CLI:npx skills add obra/superpowers
Recommended Rules
View more rules βπ Code Review Agent - Thorough & Constructive Reviewer
Agentic AICode ReviewQuality Assurance
You are an expert code review agent that provides thorough, constructive, and actionable feedback. Apply systematic reasoning to evaluate code quality...
Python Testing Best Practices
PythonTestingQuality Assurance
You are an expert in Python testing with pytest and testing best practices. Key Principles: - Write tests before or alongside code (TDD/BDD) - Aim fo...
Next.js Testing Strategies Expert
Next.jsTestingJest
You are an expert in Next.js testing strategies and best practices. Key Principles: - Test Server and Client Components differently - Use Jest for un...
Recommended Workflows
View more workflows βSetup Service Worker for Offline
PWAOfflineService Worker
--- description: Enable offline functionality --- 1. **Install Workbox**: // turbo - Run `npm install next-pwa` 2. **Configure**: ```js ...
Setup Sentry Error Tracking
MonitoringSentryErrors
--- description: Track and debug production errors with Sentry --- 1. **Install Sentry SDK**: - Install the Next.js SDK. // turbo - Run `npm...
Setup Storybook for Components
StorybookComponentsTesting
--- description: Build and test components in isolation with Storybook --- 1. **Install Storybook**: - Initialize Storybook in your project. //...