Index of every page on this site: llms.txt
All skills
rohitg00 avatar

/structured-review

@d2e5c34
by Rohit Ghumarerohitg00/skillkit1.6k stars
154
Run it yourself

Performs a structured five-stage code review covering requirements compliance, correctness, code quality, testing, and security/performance. Each stage uses targeted checklists and categorized feedback (Blocker/Major/Minor/Nit) with actionable suggestions and rationale. Use when the user asks for code review, PR feedback, pull request review, or wants their code checked for bugs, style issues, or vulnerabilities — triggered by phrases like "review my code", "check this PR", "review my changes", "pull request review", or "code feedback".

  • 1 file
  • 4.9 KB
  • Updated 9 months ago
  • GitHub
Use this Skill: https://skilld.dev/gh/rohitg00/skillkit/structured-review

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈141 tokens for metadata: the name and description. ≈1k when used: this file.

Description uses 7.1% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈567 characters, or 7.1%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Structured Code Review

You are performing a structured, multi-stage code review. This methodology ensures thorough review while providing actionable, constructive feedback.

Core Principle

Review in stages. Each stage has a specific focus. Don't mix concerns.

A structured review catches more issues and provides better feedback than an unstructured scan.

Review Stages

Stage 1: Requirements Compliance

First, verify the code meets its requirements.

Checklist:

  • Implements stated requirements
  • Handles specified edge cases
  • No scope creep (unexpected additions)
  • No missing functionality

Feedback at this stage:

  • "This doesn't appear to handle the case when X is empty"
  • "The requirement specified Y, but this implements Z"
  • "This adds feature F which wasn't requested - is that intentional?"

Stage 2: Correctness

Next, verify the code works correctly.

Checklist:

  • Logic is sound
  • No obvious bugs
  • Error paths are handled
  • No unfinished code (TODOs without tickets)

Feedback at this stage:

  • "This will throw if user is null"
  • "The loop exits early before processing all items"
  • "What happens when the API call fails?"

Stage 3: Code Quality

Then, evaluate code quality and maintainability.

Checklist:

  • Clear naming
  • Reasonable function/method length
  • No unnecessary complexity
  • Follows project conventions
  • Appropriate abstractions

Feedback at this stage:

  • "Could you rename data to userProfile for clarity?"
  • "This function is doing three things - consider splitting"
  • "We use camelCase for variables in this project"

Stage 4: Testing

Evaluate test coverage and quality.

Checklist:

  • New code has tests
  • Tests cover main paths and edge cases
  • Tests are readable and maintainable
  • Tests don't test implementation details

Feedback at this stage:

  • "Please add a test for the error case"
  • "This test will break if we change the implementation"
  • "Consider using a parameterized test for these cases"

Stage 5: Security & Performance

Finally, check for security and performance concerns.

Checklist:

  • No SQL injection, XSS, etc.
  • Secrets not exposed
  • No obvious N+1 queries
  • No unnecessary computation
  • Sensitive data handled correctly

Feedback at this stage:

  • "This input should be sanitized before use"
  • "Consider adding an index for this query"
  • "This API key should come from environment variables"

Writing Good Feedback

Feedback Levels

Level When to Use Example
Blocker Must fix before merge "Security: This allows SQL injection"
Major Should fix, but not critical "This will fail for empty arrays"
Minor Suggestion, nice to have "Consider renaming for clarity"
Nit Trivial, stylistic "Extra blank line here"

Constructive Feedback Template

[Level] [Category]: [Issue]

**What:** [Describe the specific issue]
**Why:** [Explain why it matters]
**Suggestion:** [Offer a specific improvement]

Example:

[Major] Correctness: Null reference possible

**What:** `user.email` is accessed without checking if user exists
**Why:** This will throw TypeError when user is not found
**Suggestion:** Add `if (!user) return null;` before accessing properties

Review Checklist Summary

## Review: [PR Title]

### Stage 1: Requirements
- [ ] Implements requirements
- [ ] Handles edge cases
- [ ] Appropriate scope

### Stage 2: Correctness
- [ ] Logic is sound
- [ ] No bugs
- [ ] Errors handled

### Stage 3: Quality
- [ ] Readable
- [ ] Follows conventions
- [ ] Maintainable

### Stage 4: Testing
- [ ] Has tests
- [ ] Tests are good

### Stage 5: Security/Performance
- [ ] No vulnerabilities
- [ ] No performance issues

### Verdict: [ ] Approve [ ] Request Changes [ ] Comment

Integration with Other Skills

  • planning/verification-gates: Review is a key gate
  • testing/test-patterns: Evaluate test quality
  • testing/anti-patterns: Spot testing issues

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No third-party reports yet.

Signed by skilld at d2e5c34. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last month.

Steadyupdated 9 months ago
version
1.0.0
tags
[
  "collaboration",
  "code-review",
  "quality",
  "feedback"
]
difficulty
intermediate
Other metadata
triggers
[
  "code review",
  "review this",
  "feedback on code",
  "PR review",
  "review my changes"
]
estimatedTime
15
relatedSkills
[
  "collaboration/parallel-investigation",
  "planning/verification-gates"
]

README badge

README badge for rohitg00/skillkit/structured-review