Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Jiang-Yude/skill-master --skill skill-mastergit clone --depth 1 https://github.com/Jiang-Yude/skill-masterWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/jiang-yude/skill-master/skill-master)<a href="https://agentmods.dev/skills/jiang-yude/skill-master/skill-master"><img src="https://agentmods.dev/badge/skills/jiang-yude/skill-master/skill-master/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/jiang-yude/skill-master/skill-master"><img src="https://agentmods.dev/badge/skills/jiang-yude/skill-master/skill-master.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00322 | $0.07573 |
| Opus 5.5 | $0.00129 | $0.03029 |
| Sonnet 5.5 | $0.00064 | $0.01515 |
| Haiku 4.5 | $0.00032 | $0.00757 |
Grade A, and why
skill-master scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 581 lines — stays where its author put it; the contents beside it link to each section on GitHub.
技能包大師 公開版 v1.0
中文版技能包全生命週期管理系統。整合 Anthropic 官方 skill-creator + 中文化處理 + 規格驅動開發 + 評估與優化 + 觸發調校 + 技能包寫作指引。
目標：讓你做出第一個可用的技能包，並具備測試與調校的能力。
五種工作模式總覽
| 情境 | 推薦模式 | 階段 |
|---|---|---|
| 快速做一個簡單技能包 | 模式一：引導探索型 | 建 |
| 有完整內容要整理 | 模式二：專業整理型 | 建 |
| 複雜技能包或需長期維護 | 模式三：規格驅動型 | 建 |
| 技能包做好了，要驗證品質 | 模式四：評估與優化 | 測 |
| 技能包沒被正確觸發 | 模式五：觸發調校 | 調 |
完整生命週期：建 → 測 → 調 → 再測 → 再調 → ...
技能包標準檔案結構
所有技能包一律採用以下結構：
skill-name/
├── SPEC.md ← 設計圖（規格驅動模式使用，選填）
├── SKILL.md ← 主檔案（必要）
├── agents/openai.yaml ← OpenAI / Codex 類環境的 UI metadata（建議）
├── references/ ← 參考文件（按需載入）
├── scripts/ ← 可執行腳本（確定性/重複性任務）
└── assets/ ← 模板、圖示、字型等靜態資源
SPEC.md 不是每個技能包都需要。簡單技能包直接寫 SKILL.md 就好。複雜的、會長期迭代的才需要 SPEC.md。
如果技能包要拿去不同 Agent 環境獨立運作，建議補上 agents/openai.yaml。它不是技能本體的必要條件，但能讓 OpenAI / Codex 類介面正確顯示名稱、短描述與預設提示詞。
技能包寫作指引
這一節定義「怎麼寫出好的技能包」，適用於所有模式產出的技能包。
Progressive Disclosure：三層載入
技能包用三層式載入，控制 context 佔用：
| 層級 | 內容 | 何時載入 | 長度建議 |
|---|---|---|---|
| 第一層 | name + description | 永遠在 context 裡 | 約 100 字 |
| 第二層 | SKILL.md 本體 | 技能包被觸發時 | 500 行以內（理想值，可超過但要有意識控制） |
| 第三層 | references/ 裡的文件 | 需要時才讀 | 不限，但單檔超過 300 行建議加目錄 |
設計原則：SKILL.md 放「什麼時候做什麼」的決策邏輯，references/ 放「怎麼做」的操作細節。使用者不需要每次都讀完所有 references/，SKILL.md 裡面要寫清楚「什麼時候去讀哪個檔案」。
按領域拆分 references/ 的範例：
cloud-deploy/
├── SKILL.md（流程 + 選擇邏輯）
└── references/
├── aws.md
├── gcp.md
└── azure.md
Claude 只讀跟當前任務相關的那一個。
自由度設計：不要每次都寫到一樣細
技能包不是越細越好，而是要根據任務的脆弱程度，決定要給模型多大自由度。
三種自由度：
- 高自由度：用文字原則引導。適合情境差異大、沒有唯一正解、需要現場判斷的任務。
- 中自由度：用偽代碼、模板、可調參數腳本。適合有偏好做法，但仍需依情境調整的任務。
- 低自由度：用固定腳本、固定步驟、少量參數。適合容易出錯、順序不能亂、穩定性優先的任務。
判斷原則：
- 任務像走平地，路很多都能到 → 高自由度
- 任務像有欄杆的山路，需要方向但不是唯一走法 → 中自由度
- 任務像窄橋，走錯一步就掉下去 → 低自由度
寫技能包前先問自己：這件事最怕的是「做不出來」，還是「做錯一步整個壞掉」？越怕出錯，越要收斂自由度。
先規劃可重用資源，再寫 SKILL.md
不要一開始就衝去寫 SKILL.md。先看 2-3 個真實案例，判斷哪些東西值得抽成可重用資源。
最小資源規劃表：
| 資源類型 | 什麼時候該做 | 例子 |
|---|---|---|
| scripts/ | 同樣的程式碼會重寫，或任務需要高確定性 | rotate_pdf.py、clean_csv.py |
| references/ | 細節很多，但不是每次都要載入 | schema.md、api-notes.md |
| assets/ | 最終輸出會直接用到的靜態檔案 | ppt 模板、logo、前端骨架 |
What ships with it
20 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- agents/openai.yaml 305 B
- assets/eval_review.html 6.9 KB
- references/attribution.md 5.3 KB
- references/compatibility.md 3.7 KB
- references/decision-ladder-protocol.md 3.0 KB
- references/eval-guide.md 7.6 KB
- references/schemas.md 12 KB
- references/skill-spec-template.md 14 KB
- references/skill-writing-guide.md 3.9 KB
- references/trigger-tuning-guide.md 6.8 KB
- references/why-spec-driven.md 9.1 KB
- scripts/__init__.py 0 B runs code
- scripts/aggregate_benchmark.py 14 KB runs code
- scripts/generate_report.py 13 KB runs code
- scripts/improve_description.py 10 KB runs code
- scripts/package_skill.py 4.1 KB runs code
- scripts/quick_validate.py 7.4 KB runs code
- scripts/run_eval.py 11 KB runs code
- scripts/run_loop.py 13 KB runs code
- scripts/utils.py 1.6 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 581 lines · 322 tokens per session scan A 44ad8988b41b
skill-master is a skill published in the GitHub repository Jiang-Yude/skill-master (22 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 322 tokens to every session and 7,573 once invoked, about $0.0013 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
skill-master compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| skill-masterJiang-Yude/skill-master | Skill | 22 | 3mo ago | Apache-2.0 | A | 322 | Codex |
| imagegenopenai/codex | Skill | 135k | today | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 100k | today | Apache-2.0 | A | 19 | Claude Code |
| browser-automationopenclaw/openclaw | Skill | 394k | today | MIT | A | 36 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 80k | 6d ago | MIT | A | 200 | Claude Code, Codex |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 95 | Claude Code |
| chat-pet-sprite-creationmicrosoft/vscode | Skill | 195k | today | MIT | A | 53 | Claude Code, Codex |