Communitygithub.com

技能庫存審計

用於審計 Claude 技能與指令的品質。支援僅掃描已變更技能的快速掃描模式,以及完整審查的全面庫存模式,並透過序列化子代理批次評估。

技能庫存審計 是什麼?

「技能庫存審計」是一個 Claude 的斜線指令(`/skill-stocktake`),旨在對所有 Claude 技能與指令進行品質審計,採用品質檢查清單與 AI 整體判斷。它支援兩種操作模式:「快速掃描」僅評估自上次執行以來有變更的技能,適合日常維護;「完整庫存」則對所有技能進行從頭到尾的全面審查,適合定期深度檢查。該工具會掃描全域技能目錄(`~/.claude/skills/`)以及專案層級的技能目錄(如果存在)。審計流程分為四個階段:庫存盤點、品質評估(透過通用型子代理)、摘要報表與合併建議。評估結果以 JSON 格式快取,支援斷點續傳,並提供明確的裁決(保留、改進、更新、淘汰或合併),每項裁決都須附上充分的決策依據。

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

在你喜歡的 AI 中提問

開啟一個已預先載入此 Agent Skill 的新對話。

說明文件

技能庫存審計 是做什麼的?

斜線指令 (/skill-stocktake),使用品質檢查清單 + AI 整體判斷,審計所有 Claude 技能與指令。支援兩種模式:快速掃描(僅檢查最近變更的技能)與完整庫存(全面審查)。

範圍

指令會鎖定以下相對於執行目錄的路徑:

路徑說明
~/.claude/skills/全域技能(所有專案)
{cwd}/.claude/skills/專案層級技能(若目錄存在)

在第一階段開始時,指令會明確列出哪些路徑被找到並掃描。

指定特定專案

若要包含專案層級技能,請在該專案的根目錄執行:

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

若專案沒有 .claude/skills/ 目錄,則僅評估全域技能與指令。

模式

模式觸發條件持續時間
快速掃描results.json 存在(預設)5–10 分鐘
完整庫存results.json 不存在,或 /skill-stocktake full20–30 分鐘

結果快取: ~/.claude/skills/skill-stocktake/results.json

快速掃描流程

僅重新評估自上次執行後有變更的技能(5–10 分鐘)。

  1. 讀取 ~/.claude/skills/skill-stocktake/results.json
  2. 執行:bash ~/.claude/skills/skill-stocktake/scripts/quick-diff.sh \ ~/.claude/skills/skill-stocktake/results.json (專案目錄會從 $PWD/.claude/skills 自動偵測;僅在需要時才明確傳入)
  3. 若輸出為 []:回報「自上次執行後無變更。」並停止
  4. 僅使用相同的第二階段標準重新評估那些變更過的檔案
  5. 從先前的結果中沿用未變更的技能
  6. 僅輸出差異
  7. 執行:bash ~/.claude/skills/skill-stocktake/scripts/save-results.sh \ ~/.claude/skills/skill-stocktake/results.json <<< "$EVAL_RESULTS"

完整庫存流程

第一階段 — 庫存盤點

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

該腳本會列舉技能檔案、提取 frontmatter 並收集 UTC 修改時間。 專案目錄會從 $PWD/.claude/skills 自動偵測;僅在需要時才明確傳入。 從腳本輸出中呈現掃描摘要與庫存表格:

正在掃描:
  ✓ ~/.claude/skills/         (17 個檔案)
  ✗ {cwd}/.claude/skills/    (未找到 — 僅限全域技能)
技能7 天使用次數30 天使用次數說明

第二階段 — 品質評估

啟動一個 Agent 工具子代理 (通用型代理),並提供完整的庫存清單與檢查清單:

Agent(
  subagent_type="general-purpose",
  prompt="
請根據以下檢查清單評估技能庫存。

[庫存清單]

[檢查清單]

為每個技能回傳 JSON:
{ "verdict": "Keep"|"Improve"|"Update"|"Retire"|"Merge into [X]", "reason": "..." }
"
)

子代理會讀取每個技能、套用檢查清單,並回傳每個技能的 JSON:

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

區塊指導: 為保持上下文可管理,每次子代理呼叫約處理 20 個技能。每個區塊處理後,將中間結果儲存至 results.json(status: "in_progress")。

所有技能評估完畢後:將 status 設為 "completed",然後進入第三階段。

斷點續傳偵測: 若啟動時發現 status: "in_progress",則從第一個未評估的技能繼續。

每個技能都會根據以下檢查清單進行評估:

- [ ] 已檢查與其他技能的內容重疊
- [ ] 已檢查與 MEMORY.md / CLAUDE.md 的重疊
- [ ] 已檢查技術參考資料的新鮮度(若出現工具名稱 / CLI 標誌 / API,請使用 WebSearch)
- [ ] 已考慮使用頻率

裁決標準:

裁決含義
Keep有用且符合現狀
Improve值得保留,但需要特定改進
Update引用的技術已過時(請使用 WebSearch 驗證)
Retire品質低、過時或成本效益不對稱
Merge into [X]與另一個技能有實質重疊;請指定合併目標

評估是 AI 整體判斷 — 而非數字評分表。指導維度:

  • 可操作性:包含可立即執行的程式碼範例、指令或步驟
  • 範圍契合度:名稱、觸發條件與內容一致;不會過於廣泛或狹隘
  • 獨特性:價值無法被 MEMORY.md / CLAUDE.md / 其他技能取代
  • 時效性:技術參考資料在當前環境中可正常運作

理由品質要求 — reason 欄位必須自足且有助於決策:

  • 請勿僅寫「未變更」— 務必重新陳述核心證據
  • 對於 Retire:說明 (1) 發現了什麼具體缺陷、(2) 哪個項目涵蓋了相同的需求
    • 不良:"已取代"
    • 良好:"disable-model-invocation: true 已設定;已由 continuous-learning-v2 取代,該技能涵蓋所有相同模式並加上信心評分。無任何獨特內容留存。"
  • 對於 Merge:指定目標並描述要整合哪些內容
    • 不良:"與 X 重疊"
    • 良好:"42 行的簡短內容;chatlog-to-article 的第 4 步已涵蓋相同工作流程。將「文章角度」技巧作為註解整合到該技能中。"
  • 對於 Improve:說明具體需要的變更(哪個區段、什麼操作、目標大小)
    • 不良:"太長"
    • 良好:"276 行;區段「框架比較」(L80–140) 與 ai-era-architecture-principles 重複;刪除後約可縮減至 150 行。"
  • 對於 Keep(快速掃描中僅 mtime 變更):重新陳述原始裁決理由,請勿寫「未變更」
    • 不良:"未變更"
    • 良好:"mtime 更新但內容未變更。獨特的 Python 參考資料,由 rules/python/ 明確匯入;未發現重疊。"

第三階段 — 摘要報表

技能7 天使用次數裁決理由

第四階段 — 合併

  1. 淘汰 / 合併:在向使用者確認前,針對每個檔案提供詳細的說明:
    • 發現了什麼具體問題(重疊、過時、參考中斷等)
    • 哪個替代方案涵蓋相同功能(對於 Retire:哪個現有技能/規則;對於 Merge:目標檔案及要整合的內容)
    • 移除的影響(任何相依的技能、MEMORY.md 參考或受影響的工作流程)
  2. 改進:提供具體的改進建議及理由:
    • 要變更什麼及其原因(例如「將 430 行裁剪為 200 行,因為區段 X/Y 與 python-patterns 重複」)
    • 由使用者決定是否執行
  3. 更新:呈現更新後的內容及已檢查的來源
  4. 檢查 MEMORY.md 行數;若超過 100 行,建議壓縮

結果檔案結構

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

evaluated_at:必須設為評估完成的實際 UTC 時間。 透過 Bash 取得:date -u +%Y-%m-%dT%H:%M:%SZ。切勿使用僅有日期的近似值如 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": "具體、可操作,對 X 工作流程具有獨特價值",
      "mtime": "2026-01-15T08:30:00Z"
    }
  }
}

備註

  • 評估是盲審的:所有技能(無論來源為 ECC、自撰或自動提取)均套用相同的檢查清單
  • 封存 / 刪除操作一律需要使用者明確確認
  • 不因技能來源不同而異的裁決

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.

相關技能