Code Reviewer
- 1 installs
- 13 repo stars
- Updated July 18, 2026
- ahmadulhoq/agentskel
code-reviewer is a Claude Code skill that guides an agent through a 10-point checklist for reviewing pull requests and auditing code quality.
About
code-reviewer is a Claude Code skill that walks an agent through a fixed 10-point checklist when reviewing a pull request or auditing code quality. It checks correctness, edge cases, project style, architecture layer rules, static-analysis violations, security, and tests, with platform-specific branches for Android, iOS, and Web. It also verifies dependency-upgrade approvals and flags changes to documented 'sacred behaviors'.
- 10-point review checklist covering correctness, edge cases, style, and security
- Platform-branched rules for Android (Detekt), iOS (SwiftLint), and Web (ESLint/Prettier)
- Enforces dependency-upgrade approval tiers and cross-platform parity checks
Code Reviewer by the numbers
- 1 all-time installs (skills.sh)
- Ranked #982 of 1,352 Code Review & Quality skills by installs in the Skillselion catalog
- Data as of Jul 19, 2026 (Skillselion catalog sync)
code-reviewer capabilities & compatibility
- Capabilities
- code review · security audit · dependency review
- Use cases
- code review · security audit
- Pricing
- Free
What code-reviewer says it does
When reviewing a PR, auditing code quality, or validating changes against the project's standards and sacred behaviors.
Post constructive, actionable feedback — explain why, not just what.
npx skills add https://github.com/ahmadulhoq/agentskel --skill code-reviewerAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 1 |
|---|---|
| repo stars | ★ 13 |
| Last updated | July 18, 2026 |
| Repository | ahmadulhoq/agentskel ↗ |
What it does
Guide an AI agent to review a pull request against project conventions, architecture rules, and static-analysis standards before approval.
Who is it for?
Teams wanting consistent PR reviews that check architecture layer rules, static analysis, and dependency approvals across Android, iOS, and Web.
Skip if: Automatically fixing code or running the review CI itself; it guides review judgment rather than executing tools.
When should I use this skill?
Reviewing a PR, auditing code quality, or validating changes against project standards.
What you get
A consistent, checklist-driven review that flags correctness, security, style, and dependency-approval issues.
- Structured review feedback
- Approval or block decision with rationale
By the numbers
- 10-point review checklist
Files
Code Reviewer Standards
Review Checklist
1. Correctness: Does the code do what it's supposed to? 2. Edge cases: Are error conditions handled? 3. Style: Does it follow project conventions? Check .memory/CONVENTIONS.md. 4. Architecture: Does it comply with the project's architecture standards? <!-- PLATFORM: Android --> Check .agents/standards/ANDROID_ARCHITECTURE.md. Verify module dependency rules (:feature → :data OK; no :data↔:data, no :feature↔:feature). Check UDF in ViewModels. No GlobalBus/FlowBus for cross-module events. <!-- END PLATFORM: Android --> <!-- PLATFORM: iOS --> Check .agents/standards/IOS_ARCHITECTURE.md. Verify module boundaries and dependency direction. Check MVVM/UDF patterns in ViewModels. <!-- END PLATFORM: iOS --> Check module dependency rules and layer boundaries. 5. Lint / Static Analysis: <!-- PLATFORM: Android --> CI runs Detekt on every PR and posts inline comments via Reviewdog. Check that no unresolved Detekt violations exist in new/changed lines. <!-- END PLATFORM: Android --> <!-- PLATFORM: iOS --> CI runs SwiftLint. Check that no unresolved violations exist in new/changed lines. <!-- END PLATFORM: iOS --> <!-- PLATFORM: Web --> CI runs ESLint/Prettier. Check that no unresolved violations exist in new/changed lines. <!-- END PLATFORM: Web --> If CI runs static analysis, the author must fix violations before approval. 6. Performance: Are there obvious inefficiencies? Consider UX impact. 7. Security: Does it follow security-non-negotiables rules? 8. Tests: Are there adequate tests for the changes? 9. Sacred behaviors: Does this change anything in .memory/SACRED.md? 10. Dependencies: <!-- PLATFORM: Android --> If the PR touches libs.versions.toml, gradle-wrapper.properties, or buildSrc/, <!-- END PLATFORM: Android --> <!-- PLATFORM: iOS --> If the PR touches Package.swift, Podfile, or .xcode-version, <!-- END PLATFORM: iOS --> <!-- PLATFORM: Web --> If the PR touches package.json or lock files, <!-- END PLATFORM: Web --> verify the correct upgrade tier approval was obtained per .agents/standards/DEPENDENCY_MANAGEMENT.md. Block merge if not.
Cross-Platform Impact (if blueprint configured)
If Blueprint Path is set in .memory/CONFIG.md:
- [ ] Check whether the changes touch business logic documented in
[BLUEPRINT_PATH]/specs/. If so, verify the implementation matches the spec.
- [ ] If the changes modify shared business logic (calculations, validation rules,
data contracts), flag it: a Knowledge Bus entry may be needed.
- [ ] Check
[BLUEPRINT_PATH]/parity/PARITY_MATRIX.md— does this change
affect parity status for this platform?
Skip this section if no blueprint is configured.
Documentation Standards
- Documentation must be synchronised with code changes.
- Markdown files must use consistent heading hierarchies.
- Code snippets must be executable and reflect real use cases.
- Each doc must outline: purpose, usage, parameters, and examples.
QA Procedures
- Changes involving security, system config, or agent roles require review.
- User-facing output must be clear, non-technical, and actionable.
- All major updates must include a rollback plan.
- Do not nit-pick code that matches the conventions in
.memory/CONVENTIONS.md. - Post constructive, actionable feedback — explain why, not just what.
How to Provide Feedback
- Be specific about what needs to change.
- Explain why, not just what.
- Suggest alternatives when possible.
Related skills
Improve Codebase ArchitectureSafely deepen clusters of shallow modules into cohesive, testable units while respecting their external dependencies.566k196k
Caveman ReviewGet ultra-compressed, one-line code review comments that cut noise while keeping every actionable fix.270k94.7kCodebase DesignShared vocabulary for designing deep modules: improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable.266k196kCavecrewDelegate coding tasks to specialized subagents that return compressed output, keeping the main context window usable for much longer sessions.220k94.7kCode ReviewReviews a branch or PR diff on two axes at once: conformance to coding standards plus a code-smell baseline, and whether it actually implements the original spec.201k196k
Requesting Code ReviewDispatch a consistent, high-signal code reviewer subagent that catches plan deviations and quality issues before merging or continuing development.183k264kFAQ
What does the code-reviewer skill check?
A 10-point checklist: correctness, edge cases, style, architecture, lint/static analysis, performance, security, tests, sacred behaviors, and dependency approvals.
Does it support multiple platforms?
Yes, it has platform-branched rules for Android (Detekt), iOS (SwiftLint), and Web (ESLint/Prettier).