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 xcodethink/open-claude-code-skills --skill 15-playbookgit clone --depth 1 https://github.com/xcodethink/open-claude-code-skillsWrote 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/xcodethink/open-claude-code-skills/15-playbook)<a href="https://agentmods.dev/skills/xcodethink/open-claude-code-skills/15-playbook"><img src="https://agentmods.dev/badge/skills/xcodethink/open-claude-code-skills/15-playbook/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/xcodethink/open-claude-code-skills/15-playbook"><img src="https://agentmods.dev/badge/skills/xcodethink/open-claude-code-skills/15-playbook.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 | $0.00267 | $0.01953 |
| Opus 5 | $0.00133 | $0.00977 |
| Sonnet 5 | $0.00053 | $0.00391 |
| Haiku 4.5 | $0.00027 | $0.00195 |
Grade A, and why
15-playbook 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 8d 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 — 183 lines — stays where its author put it; the contents beside it link to each section on GitHub.
指令库（Playbook）
核心定位：把历史踩坑沉淀成可复用的指令模板，让用户不用每次现场组织语言就能用上"最佳指令"。
库的位置
~/.claude/prompts/
├── README.md 库说明
├── index.yaml 元数据索引（必读，做匹配用）
├── lib/ 场景模板正文
└── modifiers/ 修饰符（可叠加到任何模板）
调用形式
| 用户输入 | 行为 |
|---|---|
/pb 或 /playbook 或 /pb list |
列出所有模板速查表 |
/pb <自然语言描述> |
模糊匹配 → 候选 → 注入完整行为契约 |
/pb <模板> <修饰符>... |
组合：模板 + 修饰符 |
/pb help |
显示用法说明 |
Claude 收到 /pb 后的行为
第 1 步：读 index.yaml
任何 /pb 调用第一件事：
读 ~/.claude/prompts/index.yaml
不要凭记忆猜模板存不存在 — 索引是真理来源。
第 2 步：判断意图
| 用户输入形式 | 判断 |
|---|---|
/pb 或 /pb list |
列表模式 |
/pb help |
帮助模式 |
/pb <文字> |
匹配模式 |
第 3 步：匹配模式 — 自然语言匹配
读 index.yaml 后，按以下顺序匹配：
- 精确匹配 id：用户输入完全等于某个
id→ 直接选定 - 触发词匹配：用户输入包含某模板的
triggers关键词 → 强候选 - 语义匹配：用户输入与模板
title或description语义相近 → 弱候选
多个候选：列出 top 3，让用户选：
找到几个相关模板：
1. **反复修不好的 bug** (02-诊断, severity: high)
- 用于：同一个 bug 修了 ≥2 次还没解决
- 触发词：反复修不好 / 改了又冒新bug
2. **审计后批量修** (02-诊断, severity: high)
- 用于：审计发现一批问题需要批量处理
- 触发词：审计发现一批问题 / 一堆问题要修
你想用哪个？或者描述更多场景细节我重新匹配。
强匹配（top 1 远远超过其他）：直接确认 + 注入，不用问。
匹配到 **反复修不好的 bug** (02-诊断)
我会按这个模板的行为契约处理。如要换模板告诉我。
[然后注入模板内容]
第 4 步：识别修饰符叠加
如果用户输入里同时出现修饰符触发词（"超谨慎" / "全量回归" / "深度思考" / "不部署" / "操作前必问"）：
- 自动叠加对应修饰符
- 在确认时明示："已叠加 修饰符A + 修饰符B"
例：
用户：/pb 部署 超谨慎 不部署
匹配：deploy-triple-guard 模板 + 超谨慎 + 不部署 修饰符
（注：超谨慎和不部署看起来矛盾 — 用户可能想说"做部署前的所有检查但先不真的部署"，确认下意图）
第 5 步：注入模板
读取匹配到的模板文件 + 修饰符文件，完整内容输出给用户（这样模板的行为契约进入对话上下文，后续我会按契约执行）。
输出格式：
【已加载模板】反复修不好的 bug + 深度思考 + 全量回归
[模板正文]
---
[修饰符正文]
---
[OK] 我已按这套指令进入工作模式。请描述具体问题。
第 6 步：列表模式
/pb 或 /pb list：输出速查表（按 category 分组）：
## 启动类
- new-feature-vertical 新功能-垂直切片
- migration-rewrite 迁移/重写启动
## 诊断类
- stuck-bug 反复修不好的 bug
- audit-batch-fix 审计后批量修
## 实施类
- spec-implementation 按 spec 做（防漂白）
- add-x-not-touch-y 加 X 不动 Y（纯新增优先）
- implement-all-clarify "全部实施"语义澄清
## 部署类
- deploy-triple-guard 部署三合一护栏
- emergency-rollback 紧急回滚
## 设计类
- ui-restore-browser-test UI 还原 + 浏览器实测
## 元指令
- lesson-distill 沉淀经验到 lesson
- hard-brake-checklist 硬刹车自检
- playbook-evolve-review Playbook 演进 review
## 修饰符（可叠加）
- deep-think 深度思考
- full-regression 全量回归
- no-deploy 不部署
- ask-before-act 操作前必问
- ultra-careful 超谨慎
用法：/pb <模板名或场景描述> [修饰符...]
例： /pb 反复修不好的bug 深度思考 全量回归
What ships with it
2 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.
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.
- 8d ago First seen · 183 lines · 267 tokens per session scan A 25e30ed141e9
15-playbook is a skill published in the GitHub repository xcodethink/open-claude-code-skills (2 stars, last pushed 1mo ago), licensed MIT. It adds 267 tokens to every session and 1,953 once invoked, about $0.0013 per session on Opus 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 2026-08-31.
Other skills, from other repositories
llm-patterns
AI-first application patterns, LLM testing, prompt management.
prompt-sensei
Stage-aware prompt coaching, prompt improvement, lookback analysis, prompting habit feedback, and local reports about prompt quality for AI coding agents such as Claude Code or Codex.
ai-evaluation-engineering
An AI evaluation engineering specialist for testing models and prompts. The description does not provide enough detail to state which concrete operations it performs.
prompt-engineering
Prompt engineering techniques and patterns. Use when writing agent commands, hooks, skills, subagent prompts, or any LLM interaction: optimizing prompts, improving output reliability, and designing production-grade prompt templates. Trigger words: prompt engineering, prompt, prompt optimization, LLM interaction.
ideogram4
Prompting patterns for Ideogram 4 text-to-image — best-in-class in-image text rendering and exact color/layout control via structured JSON captions. Use when generating images that need legible on-image text (title cards, thumbnails, logos, signage, CTAs), precise brand colors, or controlled spatial layout. Triggers…
cost-aware-llm-pipeline
A planning guide for choosing language models and managing the amount of conversation context used by an AI coding workflow. It groups tasks by complexity and gives rules for avoiding context overflow during long sessions.