Review code
with agent skills.
A useful code review connects a specific change to a concrete failure or unmet requirement. These skills help with requesting a review, inspecting a patch and working through the findings.
Filter these skills by stack and access
Choose a skill
Distinguish reviewing code from preparing a review request or responding to feedback. Choose guidance that fits the change and your stack. Keep the reviewer independent of the implementation when the risk warrants it.
Explore the review loop kitBefore you install
Name the base and head revision, the intended behavior and the areas that changed. Ask for findings with file locations, impact and supporting evidence. Say explicitly whether the task is read-only or includes making fixes.
Open a listing to read its source, access labels and review evidence before copying its install command. The automated scan covers SKILL.md, not referenced files or runtime behavior. Read the review limits.
Skills for this task
15 available skills, listed A to Z. Task labels help narrow the choice; they do not establish fit for your project.
Address GitHub comments
Finds review threads on the current branch's pull request and applies the selected fixes.
What it can touch: Edits code · Runs commands · Uses git · Network
Automated scan recorded · evidence and limits
- Automated scan: Checked 16 Nov 2026. Checks SKILL.md patterns and frontmatter; does not run the skill.
- Source revision: 49f948faa925 (opens in a new tab). The install command targets this commit. Only SKILL.md was scanned.
- Human reading: No full human-reading date recorded.
- Runtime evidence: No runtime-test evidence recorded here.
A review is not a security audit or a guarantee.
Commands target the repository commits recorded with the automated scans. Scans cover only SKILL.md, not referenced files or runtime behavior. What the review covers.
5.2Kinstalls+25 today
npx skills@latest add https://github.com/openai/skills/tree/49f948faa9258a0c61caceaf225e179651397431 --skill gh-address-commentsAudit context building
Maps what each function assumes, guarantees and depends on before a security review.
What it can touch: Subagents
Automated scan recorded · evidence and limits
- Automated scan: Checked 16 Nov 2026. Checks SKILL.md patterns and frontmatter; does not run the skill.
- Source revision: 149d7b21f406 (opens in a new tab). The install command targets this commit. Only SKILL.md was scanned.
- Human reading: No full human-reading date recorded.
- Runtime evidence: No runtime-test evidence recorded here.
A review is not a security audit or a guarantee.
Commands target the repository commits recorded with the automated scans. Scans cover only SKILL.md, not referenced files or runtime behavior. What the review covers.
7.7Kinstalls+40 today
npx skills@latest add https://github.com/trailofbits/skills/tree/149d7b21f4067765b82e67029730b42d228ef4b4 --skill audit-context-buildingCode improver
Runs a scoped reviewer-and-fixer loop until blocking findings clear or convergence fails.
What it can touch: Edits code · Runs commands · Uses git · Subagents
Automated scan recorded · evidence and limits
- Automated scan: Checked 16 Nov 2026. Checks SKILL.md patterns and frontmatter; does not run the skill.
- Source revision: 149d7b21f406 (opens in a new tab). The install command targets this commit. Only SKILL.md was scanned.
- Human reading: No full human-reading date recorded.
- Runtime evidence: No runtime-test evidence recorded here.
A review is not a security audit or a guarantee.
Commands target the repository commits recorded with the automated scans. Scans cover only SKILL.md, not referenced files or runtime behavior. What the review covers.
1.6Kinstalls+30 today
npx skills@latest add https://github.com/trailofbits/skills/tree/149d7b21f4067765b82e67029730b42d228ef4b4 --skill code-improverCode review
Sentry's engineering review checklist, with concrete patterns to flag.
What it can touch: Guidance only
Automated scan recorded · evidence and limits
- Automated scan: Checked 16 Nov 2026. Checks SKILL.md patterns and frontmatter; does not run the skill.
- Source revision: a582c51ee29d (opens in a new tab). The install command targets this commit. Only SKILL.md was scanned.
- Human reading: No full human-reading date recorded.
- Runtime evidence: No runtime-test evidence recorded here.
A review is not a security audit or a guarantee.
Commands target the repository commits recorded with the automated scans. Scans cover only SKILL.md, not referenced files or runtime behavior. What the review covers.
5.8Kinstalls+14 today
npx skills@latest add https://github.com/getsentry/skills/tree/a582c51ee29d0f35e6ac321340796f77d43da733 --skill code-reviewCode review
Reviews a diff twice in parallel: against your repo's standards and against the spec.
What it can touch: Uses git · Subagents
Automated scan recorded · evidence and limits
- Automated scan: Checked 16 Nov 2026. Checks SKILL.md patterns and frontmatter; does not run the skill.
- Source revision: 5863806dd4e1 (opens in a new tab). The install command targets this commit. Only SKILL.md was scanned.
- Human reading: No full human-reading date recorded.
- Runtime evidence: No runtime-test evidence recorded here.
A review is not a security audit or a guarantee.
Commands target the repository commits recorded with the automated scans. Scans cover only SKILL.md, not referenced files or runtime behavior. What the review covers.
896Kinstalls+2.5K today
npx skills@latest add https://github.com/mattpocock/skills/tree/5863806dd4e15c63174bee4e7515a4256792ecfa --skill code-reviewCode review and quality
A five-axis review: correctness, readability, architecture, security and performance.
What it can touch: Guidance only
Automated scan recorded · evidence and limits
- Automated scan: Checked 16 Nov 2026. Checks SKILL.md patterns and frontmatter; does not run the skill.
- Source revision: 39b723ceebf8 (opens in a new tab). The install command targets this commit. Only SKILL.md was scanned.
- Human reading: No full human-reading date recorded.
- Runtime evidence: No runtime-test evidence recorded here.
A review is not a security audit or a guarantee.
Commands target the repository commits recorded with the automated scans. Scans cover only SKILL.md, not referenced files or runtime behavior. What the review covers.
89Kinstalls+627 today
npx skills@latest add https://github.com/addyosmani/agent-skills/tree/39b723ceebf81474d0557c1fecd01eddab585563 --skill code-review-and-qualityCode reviewer
Produces a prioritized review of correctness, security, performance, architecture, and tests.
What it can touch: Guidance only
Automated scan recorded · evidence and limits
- Automated scan: Checked 16 Nov 2026. Checks SKILL.md patterns and frontmatter; does not run the skill.
- Source revision: 078a25cc5bfd (opens in a new tab). The install command targets this commit. Only SKILL.md was scanned.
- Human reading: No full human-reading date recorded.
- Runtime evidence: No runtime-test evidence recorded here.
A review is not a security audit or a guarantee.
Commands target the repository commits recorded with the automated scans. Scans cover only SKILL.md, not referenced files or runtime behavior. What the review covers.
7Kinstalls+13 today
npx skills@latest add https://github.com/jeffallan/claude-skills/tree/078a25cc5bfd8011459be7b91d004e900c1d02d3 --skill code-reviewerDjango Performance Review
Finds only validated Django ORM and request-path performance problems.
What it can touch: Runs commands · Subagents
Automated scan recorded · evidence and limits
- Automated scan: Checked 16 Nov 2026. Checks SKILL.md patterns and frontmatter; does not run the skill.
- Source revision: a582c51ee29d (opens in a new tab). The install command targets this commit. Only SKILL.md was scanned.
- Human reading: No full human-reading date recorded.
- Runtime evidence: No runtime-test evidence recorded here.
A review is not a security audit or a guarantee.
Commands target the repository commits recorded with the automated scans. Scans cover only SKILL.md, not referenced files or runtime behavior. What the review covers.
3.2Kinstalls+8 today
npx skills@latest add https://github.com/getsentry/skills/tree/a582c51ee29d0f35e6ac321340796f77d43da733 --skill django-perf-reviewIterate pull request
Repairs an open pull request until actionable CI and high- or medium-priority feedback clear.
What it can touch: Edits code · Runs commands · Uses git · Network
Automated scan recorded · evidence and limits
- Automated scan: Checked 16 Nov 2026. Checks SKILL.md patterns and frontmatter; does not run the skill.
- Source revision: a582c51ee29d (opens in a new tab). The install command targets this commit. Only SKILL.md was scanned.
- Human reading: No full human-reading date recorded.
- Runtime evidence: No runtime-test evidence recorded here.
A review is not a security audit or a guarantee.
Commands target the repository commits recorded with the automated scans. Scans cover only SKILL.md, not referenced files or runtime behavior. What the review covers.
4.7Kinstalls+8 today
npx skills@latest add https://github.com/getsentry/skills/tree/a582c51ee29d0f35e6ac321340796f77d43da733 --skill iterate-prPR review expert
Reviews pull requests for blast radius, security, breaking changes, coverage, and performance.
What it can touch: Runs commands · Uses git · Network
Automated scan recorded · evidence and limits
- Automated scan: Checked 16 Nov 2026. Checks SKILL.md patterns and frontmatter; does not run the skill.
- Source revision: 4680893d91f2 (opens in a new tab). The install command targets this commit. Only SKILL.md was scanned.
- Human reading: No full human-reading date recorded.
- Runtime evidence: No runtime-test evidence recorded here.
A review is not a security audit or a guarantee.
Commands target the repository commits recorded with the automated scans. Scans cover only SKILL.md, not referenced files or runtime behavior. What the review covers.
709installs0 today
npx skills@latest add https://github.com/alirezarezvani/claude-skills/tree/4680893d91f2aee6f77ca1519fd1b86b7a104c87 --skill pr-review-expertReceiving code review
Handles review feedback with verification instead of automatic agreement.
What it can touch: Edits code
Automated scan recorded · evidence and limits
- Automated scan: Checked 16 Nov 2026. Checks SKILL.md patterns and frontmatter; does not run the skill.
- Source revision: 22f0b463e7a5 (opens in a new tab). The install command targets this commit. Only SKILL.md was scanned.
- Human reading: No full human-reading date recorded.
- Runtime evidence: No runtime-test evidence recorded here.
A review is not a security audit or a guarantee.
Commands target the repository commits recorded with the automated scans. Scans cover only SKILL.md, not referenced files or runtime behavior. What the review covers.
241Kinstalls+1.3K today
npx skills@latest add https://github.com/obra/superpowers/tree/22f0b463e7a5e579904dd6211a520321035526f7 --skill receiving-code-reviewRequesting code review
Sends finished work to a reviewer subagent before merging.
What it can touch: Uses git · Subagents
Automated scan recorded · evidence and limits
- Automated scan: Checked 16 Nov 2026. Checks SKILL.md patterns and frontmatter; does not run the skill.
- Source revision: 22f0b463e7a5 (opens in a new tab). The install command targets this commit. Only SKILL.md was scanned.
- Human reading: No full human-reading date recorded.
- Runtime evidence: No runtime-test evidence recorded here.
A review is not a security audit or a guarantee.
Commands target the repository commits recorded with the automated scans. Scans cover only SKILL.md, not referenced files or runtime behavior. What the review covers.
283Kinstalls+1.3K today
npx skills@latest add https://github.com/obra/superpowers/tree/22f0b463e7a5e579904dd6211a520321035526f7 --skill requesting-code-reviewSecurity review
Checks application code against a practical security checklist before production work.
What it can touch: Edits code · Runs commands
Last good scan retained · evidence and limits
- Automated scan: Latest fetch failed. Last successful scan: 8 Nov 2026. Current upstream content is unverified.
- Source revision: No repository commit recorded. Install command is unpinned.
- Human reading: No full human-reading date recorded.
- Runtime evidence: No runtime-test evidence recorded here.
A review is not a security audit or a guarantee.
Installs the upstream repository’s current version, not a pinned reviewed revision. The source may have changed since the check. What the review covers.
—not counted yet
npx skills@latest add affaan-m/everything-claude-code --skill security-reviewSecurity review
Reviews code for confirmed security weaknesses with data-flow and framework context.
What it can touch: Runs commands · Subagents
Automated scan recorded · evidence and limits
- Automated scan: Checked 16 Nov 2026. Checks SKILL.md patterns and frontmatter; does not run the skill.
- Source revision: a582c51ee29d (opens in a new tab). The install command targets this commit. Only SKILL.md was scanned.
- Human reading: No full human-reading date recorded.
- Runtime evidence: No runtime-test evidence recorded here.
A review is not a security audit or a guarantee.
Commands target the repository commits recorded with the automated scans. Scans cover only SKILL.md, not referenced files or runtime behavior. What the review covers.
24Kinstalls+88 today
npx skills@latest add https://github.com/getsentry/skills/tree/a582c51ee29d0f35e6ac321340796f77d43da733 --skill security-reviewWorkers Best Practices
Write and review production Cloudflare Workers against current runtime and configuration behavior.
What it can touch: Edits code · Runs commands · Network
Automated scan recorded · evidence and limits
- Automated scan: Checked 16 Nov 2026. Checks SKILL.md patterns and frontmatter; does not run the skill.
- Source revision: cbf893b95422 (opens in a new tab). The install command targets this commit. Only SKILL.md was scanned.
- Human reading: No full human-reading date recorded.
- Runtime evidence: No runtime-test evidence recorded here.
A review is not a security audit or a guarantee.
Commands target the repository commits recorded with the automated scans. Scans cover only SKILL.md, not referenced files or runtime behavior. What the review covers.
146Kinstalls+852 today
npx skills@latest add https://github.com/cloudflare/skills/tree/cbf893b954222276bf1af57e5f23379231e0efb9 --skill workers-best-practices
Check the result
- 01
Check each finding against the current diff and its callers. Separate defects from style preferences.
- 02
Reproduce important failures where safe, and verify the proposed fix addresses the cause.
- 03
Rerun affected checks after changes. Resolve review threads only when the finding has been addressed or a decision is recorded.
Example: review a pagination change
Use this illustrative brief when a change adds pagination to an existing list. Replace the revisions, file paths and requirements with your own. This is a starting point for a review, not a report from a tested project.
Review the diff from BASE_SHA to HEAD_SHA without editing files. The list must show each matching item once, preserve the selected filter when moving between pages, and handle an empty result. Trace the changed pagination code and its callers. Report defects with a file and line, the input or sequence that triggers them, and the expected versus actual behavior. Separate confirmed findings from checks you could not perform. Do not run commands or contact services without permission.
What a useful finding would contain
For example, a reviewer might suspect that changing the filter leaves the page number unchanged. To confirm it, they would need to show the state transition in code or reproduce a case where page 4 becomes empty even though the new filter has results on page 1. A statement such as “pagination may be broken” is not enough.
What to check after a fix
Repeat the reported sequence. Check first and last pages, empty results, filter changes and repeated navigation. Record the commands or manual steps actually run. Passing this example does not establish that every pagination path is correct.