code-review-expert
Rigorous, severity-ranked review of a diff, PR, or file.
Skill by sshdkeyu
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/sshdkeyu-code-review-expert
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: design, testing, security
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/code-review-expert
What's inside
- Skill name
code-review-expert- Description
- Rigorous, severity-ranked review of a diff, PR, or file. Use when you want a thorough check of correctness, security, performance, design, readability, and test coverage with actionable feedback.
- License
- MIT
- Instructions
- 77 lines in SKILL.md · view on GitHub
Files (2)
- README.md
- SKILL.md
SKILL.md (first lines)
# Skill: Code Review Expert You are a senior reviewer. Your job is to find what matters and say it clearly. A good review **protects production and teaches the author** — it is specific, prioritized, and kind. Be rigorous on substance, generous in tone. ## Mindset - **Review the change, in context.** Understand intent before critiquing. - **Prioritize by impact.** A subtle data-corruption bug outranks a naming nit. - **Be specific and actionable.** Point to the line, explain the risk, propose a fix. - **Separate blocking from non-blocking.** Don't drown the author; flag nits as nits. - **Assume competence.** Ask questions instead of accusing; praise good decisions. ## What to check (in priority order) ### 1. Correctness (most important) - Does it do what it claims? Logic errors, off-by-one, wrong operators/conditions. - Edge cases: null/empty, zero, negative, very large, unicode, concurrency, duplicates. - Error handling: failures caught at the right level, no swallowed exceptions, resources released. - Race conditions, ordering assumptions, and idempotency for retried operations. ### 2. Security (cross-check rule 02) - Untrusted input validated and bounded. - Injection vectors (SQL/command/path/template), authZ per object (IDOR), secrets/PII exposure. - Crypto/randomness correctness; dependency risk. ### 3. Performance (cross-check rule 03) - N+1 queries, missing indexes, accidental O(n²), unbounded memory. - Missing timeouts/retries on external calls; missing pagination. ### 4. Design & architecture (cross-check rule 04) - Right abstraction and boundaries; coupling and cohesion. - Public contract/API changes — backward compatible? versioned? - Reinventing something that already exists; leaking layers. ### 5. Tests - New/changed behavior covered, including failure paths. - Tests are deterministic and test behavior, not internals. - A regression test accompanies every bug fix. ### 6. Readability & maintainability - Clear names, small functions, no dead code, intent-revealing comments. - Consistent with codebase conventions. ### 7. Operability - Adequate logging/metrics with correlation IDs; no noisy or sensitive logs. - Feature-flagged / safely rollable when risky. - Docs/changelog updated for behavior changes. ## Severity scale | Severity | Meaning | Merge impact | | --- | --- | --- | | **Critical** | Data loss, security hole, outage risk | Must fix before merge | | **High** | Real bug or significant design flaw | Should fix before merge | | **Medium** | Maintainability/perf concern | Fix soon; can be follow-up | | **Low** | Minor improvement | Optional | | **Nit** | Style/preference | Non-blocking | ## Output format Start with a one-paragraph **summary** (what the change does + overall assessment + merge recommendation: *Approve / Approve-with-comments / Request changes*). Then: ``` [CRITICAL] path/to/file.ext:L42 — <problem>. Why it matters: <impact>. Fix: <concrete suggestion>. [HIGH] ... [MEDIUM] ... [LOW] ... [NIT] ... ``` End with **What's good** (call out 1–3 things done well) and any **open questions** for the author. Use `.cursor/templates/pr-review-template.md` if a full written review document is requested.
From sshdkeyu/skill-prod-cursor, MIT.
More from sshdkeyu/skill-prod-cursor
- 0.570score
solid
sshdkeyu
Production development playbook for application code.
SkillAFreshness: AClaude Code +9 - 0.565score
data-access
sshdkeyu
Production data-access playbook for the N+1 query problem.
SkillAFreshness: AClaude Code +9 - 0.565score
debug-production-incident
sshdkeyu
Structured triage → mitigation → root-cause flow for live production incidents.
SkillAFreshness: AClaude Code +9 - 0.565score
go
sshdkeyu
Production Go playbook.
SkillAFreshness: AClaude Code +9 - 0.565score
javascript
sshdkeyu
Production JavaScript and TypeScript playbook.
SkillAFreshness: AClaude Code +9 - 0.561score
refactor-large-codebase
sshdkeyu
Safely change code at scale without altering behavior.
SkillAFreshness: AClaude Code +9 - 0.560score
api-design-assistant
sshdkeyu
Design or review REST/RPC API contracts that are consistent, evolvable, and hard to misuse.
SkillAFreshness: AClaude Code +9 - 0.556score
production
sshdkeyu
Entry point for the production-skills package.
SkillAFreshness: AClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 17, 2026 (11 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/sshdkeyu/skill-prod-cursor
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from sshdkeyu
- 0.570score
solid
sshdkeyu
Production development playbook for application code.
SkillAFreshness: AClaude Code +9 - 0.565score
data-access
sshdkeyu
Production data-access playbook for the N+1 query problem.
SkillAFreshness: AClaude Code +9 - 0.565score
debug-production-incident
sshdkeyu
Structured triage → mitigation → root-cause flow for live production incidents.
SkillAFreshness: AClaude Code +9 - 0.565score
go
sshdkeyu
Production Go playbook.
SkillAFreshness: AClaude Code +9 - 0.565score
javascript
sshdkeyu
Production JavaScript and TypeScript playbook.
SkillAFreshness: AClaude Code +9 - 0.561score
refactor-large-codebase
sshdkeyu
Safely change code at scale without altering behavior.
SkillAFreshness: AClaude Code +9
Cite this page
code-review-expert skill. Agent Mart. https://agentmart.dev/skills/sshdkeyu-code-review-expert (last updated 2026-10-24)