agentleFS
zekhoizekhoi/claude/.claude/skills/kit-code-review/SKILL.md

kit-code-review

Skill0 starsChanged 20 days agoMITView on GitHub

Review code changes for correctness, security, and maintainability defects and report evidence-based findings with severity and file locations. Use when asked to review a diff, branch, pull or merge request, commit, or specific files. Do not use for writing new features, fixing bugs, or general questions about code; this skill reports findings and does not edit code.

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    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:

    If the scope is unclear or empty, say so and ask rather than reviewing unrelated code.

    line 13
  2. Distinctive:

    Read project instructions and any project context document for conventions and commands.

    line 14
  3. Distinctive:

    Read every changed hunk and enough surrounding code to understand it: callers, callees, types, schemas, and tests.

    line 18
  4. Distinctive:

    Check, in priority order:

    line 23
  5. Distinctive:

    Verify each suspected defect before reporting it: trace the code path, check the types, or run a focused test or command when it is safe …

    line 31
  6. Very specific:

    Use: "Review my changes before I open a PR." "Can you check src/auth/session.ts for bugs?"

    line 46

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “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

Review the requested changes and report defects with evidence. Do not modify files unless the user explicitly asks for fixes after the review.

1. Establish scope

  • Identify what to review: the user's named files, commit, branch, or request; otherwise the uncommitted changes (git diff and git diff --staged), or the current branch against its base.
  • If the scope is unclear or empty, say so and ask rather than reviewing unrelated code.
  • Read project instructions and any project context document for conventions and commands.

2. Understand the change

  • Read every changed hunk and enough surrounding code to understand it: callers, callees, types, schemas, and tests.
  • Determine the intent from the description, commit messages, or code. Review against that intent.

3. Look for defects

Check, in priority order:

  1. Correctness: logic errors, wrong conditions, off-by-one, null or undefined handling, error paths, concurrency and ordering, resource cleanup, behavior on empty or large inputs.
  2. Security: injection (SQL, shell, path, template), missing authorization or validation at trust boundaries, secrets in code or logs, unsafe deserialization, data exposure.
  3. Data integrity: migrations, transactions, partial failure, backward compatibility of stored data and APIs.
  4. Tests: changed behavior without tests, tests that cannot fail, removed or weakened assertions.
  5. Maintainability: duplication of existing utilities, violations of established project conventions, misleading names or comments. Report these only when they matter; do not nitpick formatting that tooling handles.

Verify each suspected defect before reporting it: trace the code path, check the types, or run a focused test or command when it is safe and cheap. Discard findings you cannot support.

4. Report

Order findings by severity:

  • Critical: security vulnerability, data loss, or a crash in a common path.
  • High: incorrect behavior users will hit.
  • Medium: incorrect behavior in edge cases, or missing tests for risky changes.
  • Low: maintainability issues worth fixing.

For each finding give the severity, file:line, the problem, the concrete scenario that triggers it, and a suggested fix. Then list open questions, and finally anything you could not verify. If there are no findings, say so plainly and state what you checked.

Examples

  • Use: "Review my changes before I open a PR." "Can you check src/auth/session.ts for bugs?"
  • Skip: "Add pagination to the users endpoint." (feature work) "Why is this test failing?" (debugging)

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.