Back to Details

review-synthesis

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Review Synthesis

评审整合

Use this skill to run a review through several focused lenses and integrate the results into one coherent answer.
Do not spawn subagents unless the user explicitly asks for multiple agents, parallel reviewers, or named perspectives. If the user explicitly invokes this skill without explicitly allowing subagents, stop, ask one concise confirmation to spawn reviewers; and then resume as instructed. Do not require the user to rephrase. Default to read-only review unless the user explicitly asks to implement changes. This skill reviews work; 
$skill-capture
decides whether review lessons should become reusable skills.
使用此skill从多个聚焦视角开展评审，并将结果整合为一个连贯的结论。
除非用户明确要求使用多Agent、并行评审员或指定视角，否则不要生成子Agent。如果用户明确调用此skill但未明确允许生成子Agent，请暂停操作，用简洁的确认请求询问是否生成评审员，之后按指示继续。无需用户重新表述内容。默认采用只读评审模式，除非用户明确要求实施变更。此skill负责评审工作；
$skill-capture
负责决定评审经验是否应转化为可复用的skill。

Workflow

工作流程

  1. Confirm scope and mode: read-only review, implementation follow-up, or both.
  2. Inspect repo instructions, relevant docs, 
    git status --short
    , and target files enough to avoid blind delegation.
  3. If reviewing an agent skill, skill proposal, or skill collection change, read 
    references/skill-reviews.md
    and use its lenses to select reviewers and structure findings.
  4. Select relevant perspectives. Use user-named roles first; otherwise choose only roles that fit the target.
  5. Brief each reviewer with:
    • exact target files, directories, or question
    • its assigned perspective and boundaries
    • read-only instruction unless implementation was explicitly requested
    • expected output: findings first, severity, file/line references when available, and open questions
  6. Continue a lightweight parent review while reviewers run; do not duplicate their assigned work.
  7. Integrate the results:
    • deduplicate overlapping findings
    • resolve contradictions or mark them as tradeoffs
    • verify important claims against files when practical
    • separate confirmed issues from suggestions and taste calls
    • connect findings to likely user impact
  8. Present one synthesized review, not a paste-up of agent reports, with the smallest useful action path.
  1. 确认范围与模式：只读评审、落地跟进，或两者兼具。
  2. 检查仓库说明、相关文档、
    git status --short
    以及目标文件，避免盲目委托。
  3. 若评审的是Agent skill、skill提案或skill集合变更，请阅读
    references/skill-reviews.md
    ，并使用其中的视角选择评审员、梳理评审结果。
  4. 选择相关视角。优先使用用户指定的角色；否则仅选择与目标匹配的角色。
  5. 向每位评审员提供以下信息：
    • 明确的目标文件、目录或问题
    • 分配的视角与边界
    • 只读指令（除非明确要求落地实施）
    • 预期输出：先列发现问题，再说明严重程度，尽可能标注文件/行号，以及待解决问题
  6. 在评审员开展工作时，同步进行轻量化的父级评审；不要重复他们的分配任务。
  7. 整合结果：
    • 去重重叠的发现
    • 解决矛盾或标记为权衡项
    • 若可行，对照文件验证重要结论
    • 将已确认问题与建议、主观偏好区分开
    • 关联发现问题与可能对用户造成的影响
  8. 呈现一份整合后的评审报告，而非Agent报告的拼接版，同时给出最具实操性的行动路径。

Common Perspectives

常见视角

Choose lenses that fit the target:
  • Architecture: boundaries, coupling, domain model, maintainability.
  • Expert coder: correctness, idioms, tests, edge cases, best practices.
  • DevOps: deployment, CI, config, observability, reliability.
  • Security/privacy: secrets, auth, data exposure, unsafe defaults.
  • UI: implementation polish, responsiveness, accessibility mechanics.
  • UX/design: flows, clarity, hierarchy, task fit.
  • Docs/product: accuracy, reader fit, terminology, missing context.
  • Skill authoring: trigger clarity, instruction coherence, cognitive load, progressive disclosure, agent ergonomics, portability, bundled assets, and validation.
For skill-specific reviews, read 
references/skill-reviews.md
.
选择与目标匹配的视角：
  • Architecture（架构）：边界、耦合度、领域模型、可维护性。
  • Expert coder（资深开发）：正确性、编码规范、测试、边缘案例、最佳实践。
  • DevOps：部署、CI、配置、可观测性、可靠性。
  • Security/privacy（安全/隐私）：密钥、认证、数据泄露、不安全默认设置。
  • UI：实现细节、响应性、可访问性机制。
  • UX/design（UX/设计）：流程、清晰度、层级结构、任务适配性。
  • Docs/product（文档/产品）：准确性、读者适配性、术语一致性、缺失的上下文。
  • Skill authoring（Skill创作）：触发逻辑清晰度、指令连贯性、认知负荷、渐进式披露、Agent易用性、可移植性、捆绑资源、验证机制。
针对skill专项评审，请阅读
references/skill-reviews.md
。

Output Shape

输出格式

Lead with integrated findings:
text
Findings
- [severity] [perspective] File/line: issue, impact, recommendation.

Cross-Perspective Synthesis
- Consensus, tensions/tradeoffs, and parent assessment.

Open Questions
- Questions blocking a confident recommendation.

Recommended Next Actions
- Smallest useful fixes and validation.

Subagent Notes
- Brief role-by-role summary when useful.
If there are no major issues, say so clearly and list residual risks or test gaps. For code reviews, prioritize bugs, regressions, missing tests, and operational risks before style. For planning or design reviews, prioritize decision risks and unclear assumptions.
以整合后的发现开头：
text
Findings
- [severity] [perspective] File/line: issue, impact, recommendation.

Cross-Perspective Synthesis
- Consensus, tensions/tradeoffs, and parent assessment.

Open Questions
- Questions blocking a confident recommendation.

Recommended Next Actions
- Smallest useful fixes and validation.

Subagent Notes
- Brief role-by-role summary when useful.
若未发现重大问题，请明确说明，并列出剩余风险或测试缺口。对于代码评审，优先关注bug、回归问题、缺失的测试以及运维风险，其次是风格问题。对于规划或设计评审，优先关注决策风险与不明确的假设。

Guardrails

约束规则

  • Keep role briefs narrow; vague reviewers produce fog with bullet points.
  • Prefer independent lenses over several agents reading the same thing.
  • Do not treat subagent output as truth. Review, merge, and challenge it.
  • Do not bury severe findings under perspective summaries.
  • Do not make changes after a review unless the user explicitly asks for implementation.
  • Avoid decorative roles; more reviewers can mean more noise, not more insight.
  • Tie subjective design feedback to user goals, accessibility, or task fit.
  • 角色指令需明确聚焦；模糊的评审指令会产出零散无意义的要点。
  • 优先选择独立视角，而非多个Agent重复评审同一内容。
  • 不要将子Agent的输出视为绝对事实。需对其进行评审、合并与质疑。
  • 不要将严重问题隐藏在视角总结之下。
  • 除非用户明确要求实施变更，否则评审后不要进行修改。
  • 避免设置冗余角色；评审员数量过多可能产生更多噪音，而非更深入的见解。
  • 将主观设计反馈与用户目标、可访问性或任务适配性关联起来。