Skip to main content

review-commits

/code-review — Conduct full feature code review

Informações da origem

Repositório
sunpar/agentic_coding_tools
Última atividade na origem
13 de fevereiro de 2026 às 08:39
Idioma detectado do SKILL.md
inglês
Estrelas
1
Forks
0

Opções de instalação

Por padrão, está selecionado o prompt que primeiro revisa a origem. Você pode mudar para um comando direto ou baixar uma cópia local.

Revise os arquivos de origem

Leia o SKILL.md e os arquivos complementares exibidos pelo SkillsMP antes de decidir se vai instalar.

Exibindo SKILL.md

SKILL.md
Instruções da origem · Visualização somente leitura
name
review-commits
description
/code-review — Conduct full feature code review
disable-model-invocation
true
# /code-review — Conduct full feature code review You are a seasoned senior engineer performing a professional code review. Your goal is to evaluate **functionality**, **quality**, **security**, **performance**, **clarity**, and **professionalism** of the code changes in this feature. --- ## Context you should load 1. The feature's summary and intent: `@feature_summary.md` 2. The actual code changes: `@PR (Diff with Main Branch)` *(fallback: `@Commit (Diff of Working State)` or `@git` if PR not open)* --- ## Instructions Perform a detailed review using the checklist below. For each category, comment on strengths, issues, and concrete suggestions for improvement. ### Functionality - [ ] Code does what the **feature summary** says it should - [ ] Edge cases are covered - [ ] Error handling is complete and appropriate - [ ] No obvious bugs or logic errors - [ ] Behavior aligns with business intent from `feature_summary.md` ### Code Quality - [ ] Code is readable, consistent, and logically organized - [ ] Functions/classes are cohesive and single-purpose - [ ] Variable and method names are descriptive - [ ] No unnecessary complexity or duplication - [ ] Follows project conventions and style guides ### Security - [ ] No potential vulnerabilities - [ ] Input validation and sanitization are robust - [ ] Sensitive data handled securely - [ ] No hardcoded secrets or tokens - [ ] Dependencies updated and trusted ### Performance - [ ] Algorithms and data structures are efficient and appropriate - [ ] Avoids unnecessary computation, I/O, or memory allocations - [ ] Caching or memoization considered where relevant - [ ] Database queries optimized (indexes, joins, pagination) - [ ] Concurrency and async patterns used safely - [ ] Consider scalability implications (O-notation, memory footprint) ### Clarity & Professionalism - [ ] Code reads naturally and would be clear to a new team member - [ ] Comments explain "why," not "what" - [ ] Commits are well-scoped and messages meaningful - [ ] No commented-out code or debug leftovers - [ ] Documentation and naming align with professional standards ### Maintainability & Testing - [ ] Tests cover new and critical paths - [ ] Mocking/stubs used appropriately - [ ] Test names are descriptive and follow convention - [ ] Any gaps identified in unit, integration, or regression tests --- ## Output Format Produce a review summary with: **1. Executive Assessment** — 1–2 paragraphs summarizing code health, clarity, and readiness to merge. **2. Category Feedback** — A bulleted list under each checklist category summarizing findings and suggestions. **3. Priority Recommendations** — Top 3 changes that would most improve quality or performance. **4. Overall Verdict** — "Ready to merge," "Needs minor cleanup," or "Needs major revisions," with justification. --- ## Style - Be concrete and reference specific lines, symbols, or files when possible. - Ground feedback in `feature_summary.md` intent. - Be professional and constructive—aim to help the code reach production quality. Now, conduct the review.
Ver no GitHub