agentleFS
4444J994444J99/a-i--skills/skills/development/code-review-checklist/SKILL.md

code-review-checklist

Skill17 starsChanged 5 months agoApache-2.0Archived repositoryView on GitHub

Guide an AI agent through a structured code review — correctness, security, performance, readability, and test coverage — producing a summary with actionable, prioritized feedback. Triggers on pull-request review, diff review, or "review this code" requests.

29 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use intelligent-routing, nextjs-react-expert and 7 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • SKB3002Skill

    intelligent-routing

    Automatic agent selection and intelligent task routing. Analyzes user requests and automatically selects the best specialist agent(s) without requiring explicit user mentions.

    Also in 5 projects that use this

  • SKB3002Skill

    nextjs-react-expert

    React and Next.js performance optimization from Vercel Engineering. Use when building React components, optimizing performance, eliminating waterfalls, reducing bundle size, reviewing code for performance issues, or implementing server/client-side optimizations.

    Also in 3 projects that use this

  • xfstudioSkill

    rust-pro

    Master Rust 1.75+ with modern async patterns, advanced type system

    Also in 5 projects that use this

  • jornalistainclusivoSkill

    design-spec

    How to author a DESIGN.md file — the machine-readable design-token + human-rationale format that must exist before any UI is built. YAML front-matter token schema (colors, typography, spacing, rounded, components), type system, token references, and canonical section order.

    Also in 3 projects that use this

  • zzfancitizenSkill

    verify-changes

    Verify code changes do not break the project by running tests, linter checks, and import validation. Use after making any code changes, editing Python files, modifying imports, updating dependencies, or refactoring.

    Also in 3 projects that use this

  • VercelSkill

    web-design-guidelines

    Review UI code for Web Interface Guidelines compliance. Use when asked to "review my UI", "check accessibility", "audit design", "review UX", or "check my site against best practices".

    Also in 6 projects that use this

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 15 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Do error paths actually handle errors, or just swallow/log and continue silently?

    line 43
  2. Distinctive:

    If a bug is being fixed, is there a regression test that would have caught it — i.e.

    line 87
  3. Very specific:

    Do existing tests still pass, and did the change quietly weaken an assertion to make them pass (e.g.

    line 89
  4. Distinctive:

    Never comment on formatting a linter/formatter already enforces automatically.

    line 115

“Common” is measured: kept by at least 6 in 10 of the 15 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Code Review Checklist

A structured checklist for reviewing code changes — whether a diff, a pull request, or a single file — that keeps feedback consistent, prioritized, and actionable instead of scattered or stylistic-only.

When to Use

  • Reviewing a pull request or diff before merge
  • A human or another agent asks "review this code"
  • Self-review before opening a PR
  • Auditing an existing file for latent issues

The Five Dimensions

Work through each dimension in order. Skip a dimension only if it is genuinely inapplicable (e.g. "performance" for a docs-only change) — say so explicitly rather than silently omitting it.

1. Correctness

  • Does the change do what it claims to do? Trace the logic against the stated intent (PR description, issue, commit message).
  • Are edge cases handled: empty input, null/None, zero, negative numbers, empty collections, concurrent access, off-by-one boundaries?
  • Do error paths actually handle errors, or just swallow/log and continue silently?
  • Does it introduce a regression? Diff against prior behavior, not just the new code in isolation.
  • Are there any logic inversions (== vs !=, && vs ||, inclusive vs exclusive bounds)?

2. Security

  • Untrusted input: is anything from a user, network request, file, or environment variable interpolated into a shell command, SQL query, HTML template, or file path without sanitization?
  • Secrets: any hardcoded credentials, tokens, or keys — including in test fixtures or comments?
  • Authorization: does the change check permissions/ownership before acting, or does it trust a client-supplied ID?
  • Dependency risk: does it add a new third-party dependency, and if so, is it maintained and reasonably scoped?
  • Injection classes to scan for explicitly: command injection, SQL injection, path traversal, SSRF, log injection (unsanitized values written to logs an operator later greps/pastes), and unpinned/mutable CI Action tags used as a supply-chain vector.

3. Performance

  • Any obviously quadratic-or-worse operation on a collection that could be large in production (nested loops over the same dataset, repeated linear scans)?
  • N+1 query patterns (a loop issuing one DB/API call per iteration instead of batching)?
  • Unbounded growth: does the change introduce a cache, buffer, or accumulator with no eviction or size cap?
  • Are expensive operations (network calls, disk I/O, crypto) on a hot path that could be memoized, batched, or moved off it?

4. Readability & Maintainability

  • Can a stranger to this code understand the change from the diff and its names alone, without needing the author's context?
  • Is the change scoped to one concern, or does it mix refactoring with new behavior (making the diff harder to review and revert)?
  • Are names accurate and specific (not data, temp, handleStuff)?
  • Is there dead code, commented-out blocks, or leftover debug output (console.log, print, pdb.set_trace()) left in?
  • Does it follow the file's/repo's existing conventions (formatting, error-handling style, module layout) rather than introducing a new one inconsistently?

5. Test Coverage

  • Does the change include tests for the new behavior, or only for the happy path?
  • Are the edge cases identified in "Correctness" above actually covered by a test, not just mentioned in a comment?
  • If a bug is being fixed, is there a regression test that would have caught it — i.e. does the new test fail against the old code and pass against the new code?
  • Do existing tests still pass, and did the change quietly weaken an assertion to make them pass (e.g. loosening a strict equality check, or adding a broad try/except: pass)?

Output Format

Produce a summary, not a transcript of every line examined:

## Code Review: <file/PR>

**Correctness**: <finding or "no issues found">
**Security**: <finding or "no issues found">
**Performance**: <finding or "no issues found">
**Readability**: <finding or "no issues found">
**Test Coverage**: <finding or "no issues found">

### Priority Actions
1. <highest-priority fix, with file:line>
2. <next priority>
...
  • Order findings by actual risk/impact, not by which dimension they came from.
  • Cite file:line for every finding — a review comment without a location is not actionable.
  • Distinguish a blocking issue (correctness bug, security hole, missing critical test) from a suggestion (style, minor readability nit) — do not let nits drown out real defects.
  • Never comment on formatting a linter/formatter already enforces automatically.

Anti-Patterns to Avoid

  • Rubber-stamping: approving without tracing at least the changed logic paths against their stated intent.
  • Style-only reviews: flagging naming/formatting while missing a real correctness or security defect in the same diff.
  • Vague feedback: "this could be cleaner" without a concrete alternative.
  • Reviewing in isolation: judging a diff hunk without reading enough surrounding context to know whether it's actually correct in situ.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.