code-review
/code-review
Run a full pre-release review of a project - automated checks, independent reviewer, security reviewer, store-compliance checklist, and a manual test script. Use when the user asks to review code, make a release bug-free, or check before shipping.
- Author
- yuvarajsalla20 ↗
- Repository
- code-review-kit ↗
- Licence
- MIT
- Last repository change
- 06.11.2026
Assessment and security
- Files in this package
- 1
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 06.11.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
Run a full pre-release review of a project - automated checks, independent reviewer, security reviewer, store-compliance checklist, and a manual test script. Use when the user asks to review code, make a release bug-free, or check before shipping.
Original author description
Run a full pre-release review of a project - automated checks, independent reviewer, security reviewer, store-compliance checklist, and a manual test script. Use when the user asks to review code, make a release bug-free, or check before shipping.
How it works
- Steps
- Notes
SKILL.md preview
--- name: code-review description: Run a full pre-release review of a project - automated checks, independent reviewer, security reviewer, store-compliance checklist, and a manual test script. Use when the user asks to review code, make a release bug-free, or check before shipping. --- # Code review kit Findings first, fixes only after the owner chooses. Never change code during the review. ## Steps 1. **Config.** Look for `review.config.json` in the project root. If missing, copy `templates/review.config.json`, fill it from the project (stack, commands, risky areas), and show it to the owner. 2. **Automated checks.** Run `node <kit>/scripts/check.mjs` from the project root. It runs the configured commands, scans for leftover dev code and secrets, and, when the config has a `structure` block, checks for import loops, layer breaks, repeated code and oversized files, and, with a `speed` block, speed candidates. Save the output. For the full duplicate and structure table, run the `structure-review` skill. 3. **Independent review.** Start the `reviewer` agent in a fresh context. Give it only the project path, the config, the kit path and the check output, not your own opinions of the code. 4. **Security review.** Start the `security-reviewer` agent the same way. 4b. **Failure paths (on request, or automatically when the change touches `riskyAreas`).** Say "review with negative cases" to turn it on. Start a fresh `reviewer` (with the kit path) and `checklists/failure-paths.md` plus the project's `failurePaths` notes from the config. It reports only verified findings, lists what it handled correctly, and lists what needs a device or emulator. After any fixes, run it again on the fix diff: fixes open new failure paths. It is slower and costs more than the normal pass, so it stays off for small changes. 5. **Store compliance.** If `config.stores` is set, run `node <kit>/scripts/release.mjs --report release-report.md`, then start the `release-reviewer` agent the same way with that report (it walks `checklists/app-store.md` and `release.md`). 5b. **Speed (on by default).** When the config has a `speed` block, run `node <kit>/scripts/perf.mjs --report speed-report.md`. Start the `performance-reviewer` agent the same way, with that report if there is one; it reads `design.kind` (`app` or `library`) and opens every script row before keeping it. The owner can say "skip speed" for a small change. 5c. **Optional passes.** On request, or when the change touches `riskyAreas`, run the `focused-review` skill for tests (`test-reviewer`) or usability (`ux-reviewer`). 6. **Merge.** De-duplicate, verify each Blocker and High yourself by reading the cited code, drop anything you cannot confirm, and write one report using `templates/report.md`. 7. **Show the owner** the ranked list and ask which to fix. Fix only those, then re-run steps 2 to 5. 8. **Manual test script.** Produce a short numbered list of on-device steps from `templates/manual-test.md`, adjusted to the app's screens. ## Notes - Be honest about what cannot be checked (real devices, live backend). Never promise zero bugs. - Keep `review.config.json` the only project-specific file. Do not edit the kit for one project; add a checklist instead.
Versions
| Repository | Score | Licence |
|---|---|---|
| This page · yuvarajsalla20/code-review-kit | 90 | MIT |
| microsoft/aspire | 100 | MIT |
| kdlbs/kandev | 100 | AGPL-3.0 |
| dotnet/runtime | 100 | MIT |
| drosam/agents | 100 | MIT |
| ggml-org/llama.cpp | 100 | MIT |
| ShamelesAbyss/Nyx | 100 | Apache-2.0 |
| dzhng/jevgrep | 100 | MIT |
| iuliandita/skills | 100 | MIT |
Database snapshot: 13.11.2026 · package version fdd32c0247cd (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 100unified-review
Orchestrates multi-domain review (code, arch, tests, security) in a single pass. Use when thorough pre-release review is needed. - 90test-plan
Create a full test plan or test strategy document for a feature, release, or project. Use when the user asks for a test plan, test strategy, or QA planning for a release. - 90audit
Audit current changes, a path, or the full project for quality, security, performance, or test problems and record durable findings. Independent mode prepares or completes a fresh-reviewer checkpoint handoff. Use for /audit, independent review, security review, code quality… - 100pre-commit-review
Review staged and unstaged changes before committing. Use when asked for a pre-commit review, a code-quality review of the current diff, or when a solve workflow requests an independent review. - 100code-review-checklist
Review checklist - correctness, security, performance, quality, tests, AI code. Use when reviewing code or a PR. - 100code-reviewer
Code review, implementation review, 源码评审、实现复审。Use when: implementation Candidate 已形成,需要基于已批准需求/设计和精确 Git 边界做首次完整 Code Review 或整改 delta 复审。Do not use for API/HLD/LLD/Test/Runbook review or deployment approval.