Communitygithub.com

技能审查

在审查 Claude 技能和命令质量时使用。支持快速扫描(仅审查已更改的技能)和全面盘点两种模式,通过顺序子代理批量评估。

技能审查 是什么?

这是一个用于审计 Claude 技能和命令质量的 AI 代理技能。它提供了一个结构化的工作流程,评估技能是否符合质量检查清单,并基于人工智能的整体判断做出决策。该工具支持两种操作模式:快速扫描模式用于快速重新评估自上次运行以来发生变化的技能(约5-10分钟),全面盘点模式则对所有技能进行从头到尾的完整审查(约20-30分钟)。关键亮点包括基于缓存的增量扫描、阶段式工作流程(清单、评估、汇总、整合)、在长时间运行中可恢复的进度持久化,以及一个结构化的结果文件模式。

兼容平台✓Claude Code~Codex CLI~Cursor
npx skills add https://github.com/affaan-m/everything-claude-code/tree/main/skills/skill-stocktake

在你喜欢的 AI 中提问

打开一个已预加载此 Agent Skill 的新对话。

文档

技能审查 是做什么的?

斜杠命令(/技能审查),使用质量检查清单 + AI 整体判断来审计所有 Claude 技能和命令。支持两种模式:快速扫描用于最近更改的技能,全面盘点用于完整审查。

范围

该命令针对以下路径(相对于命令执行目录):

路径描述
~/.claude/skills/全局技能(所有项目)
{cwd}/.claude/skills/项目级技能(如果该目录存在)

在阶段 1 开始前,命令会明确列出已找到并扫描的路径。

定位到特定项目

要包括项目级技能,请从项目根目录运行:

cd ~/path/to/my-project
/skill-stocktake

如果项目没有 .claude/skills/ 目录,则仅评估全局技能和命令。

模式

模式触发条件持续时间
快速扫描results.json 存在(默认)5–10 分钟
全面盘点results.json 缺失,或 /skill-stocktake full20–30 分钟

结果缓存: ~/.claude/skills/skill-stocktake/results.json

快速扫描流程

仅重新评估自上次运行以来发生变化的技能(5–10 分钟)。

  1. 读取 ~/.claude/skills/skill-stocktake/results.json
  2. 运行:bash ~/.claude/skills/skill-stocktake/scripts/quick-diff.sh \ ~/.claude/skills/skill-stocktake/results.json (项目目录自动从 $PWD/.claude/skills 检测;仅在需要时显式传递)
  3. 如果输出为 []:报告“自上次运行以来无更改”并停止
  4. 仅使用相同的阶段 2 标准重新评估那些已更改的文件
  5. 保留上一轮结果中未更改的技能
  6. 仅输出差异
  7. 运行:bash ~/.claude/skills/skill-stocktake/scripts/save-results.sh \ ~/.claude/skills/skill-stocktake/results.json <<< "$EVAL_RESULTS"

全面盘点流程

阶段 1 — 清单

运行:bash ~/.claude/skills/skill-stocktake/scripts/scan.sh

该脚本枚举技能文件,提取 frontmatter,并收集 UTC mtime。 项目目录自动从 $PWD/.claude/skills 检测;仅在需要时显式传递。 显示脚本输出中的扫描摘要和清单表格:

正在扫描:
  ✓ ~/.claude/skills/         (17 个文件)
  ✗ {cwd}/.claude/skills/    (未找到 — 仅全局技能)
技能7 天使用30 天使用描述

阶段 2 — 质量评估

启动一个 Agent 工具子代理(通用代理),包含完整的清单和检查清单:

Agent(
  subagent_type="general-purpose",
  prompt="
根据以下检查清单评估技能清单。

[INVENTORY]

[CHECKLIST]

为每个技能返回 JSON:
{ "verdict": "Keep"|"Improve"|"Update"|"Retire"|"Merge into [X]", "reason": "..." }
"
)

子代理读取每个技能,应用检查清单,并返回每个技能的 JSON:

{ "verdict": "Keep"|"Improve"|"Update"|"Retire"|"Merge into [X]", "reason": "..." }

分块指南: 每次子代理调用处理约 20 个技能,以保持上下文可管理。每处理完一个块后,将中间结果保存到 results.json(status: "in_progress")。

所有技能评估完成后:设置 status: "completed",进入阶段 3。

恢复检测: 如果启动时发现 status: "in_progress",则从第一个未评估的技能处恢复。

每个技能根据以下清单进行评估:

- [ ] 已检查与其他技能的内容重叠
- [ ] 已检查与 MEMORY.md / CLAUDE.md 的重叠
- [ ] 已验证技术参考的新鲜度(如果存在工具名称/CLI 标志/API,使用 WebSearch)
- [ ] 考虑了使用频率

裁决标准:

裁决含义
Keep有用且当前有效
Improve值得保留,但需要特定的改进
Update引用的技术已过时(用 WebSearch 验证)
Retire质量低、过时或成本不对称
Merge into [X]与另一个技能有实质重叠;指明合并目标

评估是整体性 AI 判断 — 而非数字评分标准。指导维度:

  • 可操作性:可以立即执行的代码示例、命令或步骤
  • 范围契合度:名称、触发器和内容保持一致;不过宽或过窄
  • 独特性:价值不可被 MEMORY.md / CLAUDE.md / 其他技能替代
  • 时效性:技术引用在当前环境中有效

原因质量要求 — reason 字段必须自成体系且能支持决策:

  • 不要单独写“未更改” — 始终重述核心证据
  • 对于 Retire:说明 (1) 发现了什么具体缺陷,(2) 什么替代方案覆盖了相同的需求
    • 错误:"已取代"
    • 正确:"disable-model-invocation: true 已设置;已被 continuous-learning-v2 取代,该技能覆盖了所有相同模式并增加了置信度评分。没有剩余独特内容。"
  • 对于 Merge:指明目标并描述要整合的内容
    • 错误:"与 X 重叠"
    • 正确:"42 行单薄内容;chatlog-to-article 的步骤 4 已经覆盖相同的工作流。将该‘文章角度’提示作为备注整合到该技能中。"
  • 对于 Improve:描述所需的具体更改(什么部分、什么操作、相关目标大小)
    • 错误:"太长"
    • 正确:"276 行;‘框架比较’部分 (L80–140) 重复了 ai-era-architecture-principles;删除它以缩短至约 150 行。"
  • 对于 Keep(快速扫描中仅 mtime 更改):重述原始裁决理由,不要写“未更改”
    • 错误:"未更改"
    • 正确:"mtime 已更新但内容未变。独特的 Python 引用被 rules/python/ 显式导入;未发现重叠。"

阶段 3 — 汇总表

技能7 天使用裁决原因

阶段 4 — 整合

  1. Retire / Merge:在确认前,每个文件提供详细的依据:
    • 发现的具体问题(重叠、过时、引用错误等)
    • 覆盖相同功能的替代方案(对于 Retire:哪个现有技能/规则;对于 Merge:目标文件和要整合的内容)
    • 删除的影响(任何依赖的技能、MEMORY.md 引用或受影响的工作流)
  2. Improve:提供具体的改进建议及理由:
    • 要更改的内容及原因(例如,“将 430 行缩减至 200 行,因为 X/Y 部分重复了 python-patterns”)
    • 用户决定是否执行
  3. Update:提供已更新的内容及已检查的引用来源
  4. 检查 MEMORY.md 行数;如果超过 100 行,建议压缩

结果文件模式

~/.claude/skills/skill-stocktake/results.json:

evaluated_at:必须设置为评估完成时的实际 UTC 时间。 通过 Bash 获取:date -u +%Y-%m-%dT%H:%M:%SZ。切勿使用仅包含日期的近似值,如 T00:00:00Z。

{
  "evaluated_at": "2026-02-21T10:00:00Z",
  "mode": "full",
  "batch_progress": {
    "total": 80,
    "evaluated": 80,
    "status": "completed"
  },
  "skills": {
    "skill-name": {
      "path": "~/.claude/skills/skill-name/SKILL.md",
      "verdict": "Keep",
      "reason": "具体、可操作、对 X 工作流具有独特价值",
      "mtime": "2026-01-15T08:30:00Z"
    }
  }
}

注意事项

  • 评估是盲测:无论技能来源如何(ECC、自创、自动提取),均适用相同的检查清单
  • 归档/删除操作始终需要用户明确确认
  • 不根据技能来源进行裁决分支

Individual skills in this repo

This repo contains 20 individual skills — each has its own dedicated page.

accessibility

Design, implement, and audit inclusive digital products using WCAG 2.2 Level AA. Use when building or auditing UI that must meet WCAG 2.2 Level AA, or when reviewing a change for keyboard, contrast, or screen-reader support.

affaan-m/claude-api

Anthropic Claude API patterns for Python and TypeScript. Covers Messages API, streaming, tool use, vision, extended thinking, batches, prompt caching, and Claude Agent SDK. Use when building applications with the Claude API or Anthropic SDKs.

affaan-m/everything-claude-code

End-to-end marketing campaign planning and execution. Covers audience research, positioning, campaign angle definition, landing page copy, email sequences, social posts, ad copy, short-form video scripts, and content calendars. Use as the orchestration layer for multi-channel product launches. Use when planning or executing a multi-channel product launch, or producing landing page, email, social, or ad copy.

affaan-m/everything-claude-code

Development conventions and patterns for everything-claude-code. JavaScript project with conventional commits.

affaan-m/everything-claude-code-conventions

Development conventions and patterns for everything-claude-code. JavaScript project with conventional commits.

affaan-m/frontend-design

Create distinctive, production-grade frontend interfaces with high design quality. Use when the user asks to build web components, pages, or applications and the visual direction matters as much as the code quality.

affaan-m/gget

gget CLI and Python workflow for quick genomic database queries, sequence lookup, BLAST-style searches, enrichment checks, and reproducible bioinformatics evidence logs.

affaan-m/literature-review

Systematic literature-review workflow for academic, biomedical, technical, and scientific topics, including search planning, source screening, synthesis, citation checks, and evidence logging.

affaan-m/motion-ui

Production-ready UI motion system for React/Next.js. Use when implementing animations, transitions, or motion patterns.

affaan-m/project-guidelines-example

Example project-specific skill template based on a real production application.

affaan-m/pubmed-database

Direct PubMed and NCBI E-utilities search workflows for biomedical literature, MeSH queries, PMID lookup, citation retrieval, and API-backed literature monitoring.

affaan-m/scholar-evaluation

Structured scholarly-work evaluation for papers, proposals, literature reviews, methods sections, evidence quality, citation support, and research-writing feedback.

affaan-m/uspto-database

USPTO patent and trademark data workflow for official record lookup, PatentSearch queries, TSDR checks, assignment data, and reproducible IP research logs.

agent-architecture-audit

Full-stack diagnostic for agent and LLM applications. Audits the 12-layer agent stack for wrapper regression, memory pollution, tool discipline failures, hidden repair loops, and rendering corruption. Produces severity-ranked findings with code-first fixes. Essential for developers building agent applications, autonomous loops, or any LLM-powered feature. Use when an agent or LLM feature misbehaves and the failing layer is unknown, or before shipping an agent stack.

agent-eval

Head-to-head comparison of coding agents (Claude Code, Aider, Codex, etc.) on custom tasks with pass rate, cost, time, and consistency metrics. Use when choosing between coding agents, or when a change to an agent setup needs measured pass rate, cost, and time rather than an impression.

agent-harness-construction

Design and optimize AI agent action spaces, tool definitions, and observation formatting for higher completion rates. Use when defining or revising an agent

agentic-engineering

Operate as an agentic engineer using eval-first execution, decomposition, and cost-aware model routing. Use when planning or executing engineering work that agents will carry out end to end.

agentic-os

Build persistent multi-agent operating systems on Claude Code. Covers kernel architecture, specialist agents, slash commands, file-based memory, scheduled automation, and state management without external databases. Use when building a persistent multi-agent system on Claude Code with its own memory, commands, and scheduling.

agent-introspection-debugging

Structured self-debugging workflow for AI agent failures using capture, diagnosis, contained recovery, and introspection reports. Use when an agent run fails and you need a reproducible diagnosis instead of a retry.

agent-payment-x402

Add x402 payment execution to AI agents with per-task budgets, spending controls, and non-custodial wallets. Supports Base through agentwallet-sdk and X Layer through OKX Payments / OKX Agent Payments Protocol. Use when an agent must pay for something itself and needs per-task budgets, spending controls, and a non-custodial wallet.

相关技能