{"id":"a2c0c380-4ccb-45b4-8f78-2d8c162a3e70","slug":"clawhub-ldxs001-skill-standardization","name":"skill-standardization","description":"Skill 标准化规范引擎。支持 R-01~R-26 规范审查（audit / create / update / refactor / bump / readonly 六模式），含权限扫描、数据目录合规检查、渐进式加载、LLM 二次筛分类。","canonicalUrl":"https://www.xpersona.co/agent/clawhub-ldxs001-skill-standardization","sourceUrl":"https://clawhub.ai/ldxs001/skill-standardization","homepage":"https://clawhub.ai/ldxs001/skills/skill-standardization","source":"CLAWHUB","vendor":{"slug":"clawhub","label":"Clawhub","url":"https://clawhub.ai/ldxs001/skills/skill-standardization"},"protocols":["OPENCLEW"],"capabilities":[],"trustScore":null,"trustConfidence":"unknown","artifactCount":0,"benchmarkCount":0,"lastRelease":"2.103.0","freshnessAt":"2026-10-09T10:13:54.570Z","freshnessLabel":"Oct 9, 2026","securityReviewed":true,"openapiReady":false,"stats":[{"label":"Trust score","value":"Unknown"},{"label":"Compatibility","value":"OpenClaw"},{"label":"Freshness","value":"Oct 9, 2026"},{"label":"Vendor","value":"Clawhub"},{"label":"Artifacts","value":"0"},{"label":"Benchmarks","value":"0"},{"label":"Last release","value":"2.103.0"}],"factsPreview":[{"factKey":"vendor","category":"vendor","label":"Vendor","value":"Clawhub","href":"https://clawhub.ai/ldxs001/skills/skill-standardization","sourceUrl":"https://clawhub.ai/ldxs001/skills/skill-standardization","sourceType":"profile","confidence":"medium","observedAt":"2026-10-09T10:13:54.570Z","isPublic":true},{"factKey":"protocols","category":"compatibility","label":"Protocol compatibility","value":"OpenClaw","href":"https://www.xpersona.co/api/v1/agents/clawhub-ldxs001-skill-standardization/contract","sourceUrl":"https://www.xpersona.co/api/v1/agents/clawhub-ldxs001-skill-standardization/contract","sourceType":"contract","confidence":"medium","observedAt":"2026-10-09T10:13:54.570Z","isPublic":true},{"factKey":"traction","category":"adoption","label":"Adoption signal","value":"3K downloads","href":"https://clawhub.ai/ldxs001/skill-standardization","sourceUrl":"https://clawhub.ai/ldxs001/skill-standardization","sourceType":"profile","confidence":"medium","observedAt":"2026-10-09T10:13:54.570Z","isPublic":true},{"factKey":"latest_release","category":"release","label":"Latest release","value":"2.103.0","href":"https://clawhub.ai/ldxs001/skill-standardization","sourceUrl":"https://clawhub.ai/ldxs001/skill-standardization","sourceType":"release","confidence":"medium","observedAt":"2026-08-04T05:07:07.397Z","isPublic":true},{"factKey":"handshake_status","category":"security","label":"Handshake status","value":"UNKNOWN","href":"https://www.xpersona.co/api/v1/agents/clawhub-ldxs001-skill-standardization/trust","sourceUrl":"https://www.xpersona.co/api/v1/agents/clawhub-ldxs001-skill-standardization/trust","sourceType":"trust","confidence":"medium","observedAt":null,"isPublic":true}],"highlights":["3K downloads","Trust evidence available"],"agentCard":{"name":"skill-standardization","description":"Skill 标准化规范引擎。支持 R-01~R-26 规范审查（audit / create / update / refactor / bump / readonly 六模式），含权限扫描、数据目录合规检查、渐进式加载、LLM 二次筛分类。","source":"CLAWHUB","sourceId":"clawhub:s1794v2r46s8y5d1r4jdd7ec5h84rw70:skill-standardization","homepage":"https://clawhub.ai/ldxs001/skills/skill-standardization","repository":"https://clawhub.ai/ldxs001/skill-standardization","documentation":"https://www.xpersona.co/agent/clawhub-ldxs001-skill-standardization","protocols":["OPENCLEW"],"examples":[{"kind":"example","language":"text","snippet":"File v2.103.0:references/blueprint_flow.md\n\n# 蓝皮书与审计流程规范\r\n\r\n> 定义蓝皮书（Blueprint）、审计（Audit）、修复循环（Fix Loop）三者的关系与流程。\r\n> 适用于 create / update / refactor 三种模式。\r\n\r\n---\r\n\r\n## 1. 蓝皮书的精确定义\r\n\r\n**蓝皮书 = `inspect_skill()` 输出的技能目录结构快照。**\r\n\r\n### 包含内容\r\n\r\n| 字段 | 说明 | 来源 |\r\n|------|------|------|\r\n| `structure_tree` | tree 命令格式的目录树 | 现场扫描 |\r\n| `root_files` | 根目录所有文件 | `os.listdir(skill_dir)` |\r\n| `root_dirs` | 根目录所有子目录 | `os.listdir(skill_dir)` |\r\n| `ref_files` | references/ 下 .md 文件 | `os.listdir(references/)` |\r\n| `script_files` | scripts/ 下所有文件 | `os.walk(scripts/)` |\r\n| `has_data_dir` | 是否存在 data/ | `os.path.isdir(data/)` |\r\n| `has_output_dir` | 是否存在 output/ | `os.path.isdir(output/)` |\r\n| `has_logs_dir` | 是否存在 logs/ | `os.path.isdir(logs/)` |\r\n| `has_temp_dir` | 是否存在 temp/ | `os.path.isdir(temp/)` |\r\n| SKILL.md 解析 | frontmatter + H2 章节 | 文件解析 |\r\n| `_meta.json` | 元数据 | 文件解析 |\r\n| `doc_code_gaps` | 文档与代码的函数签名差异 | 代码扫描 |\r\n\r\n### 蓝皮书的合法用途（仅此两条）\r\n\r\n1. **给 LLM 看** — 帮助 LLM 理解技能结构，确定更新范围\r\n2. **给审计提供路径地图** — 审计知道该扫哪些目录（不做判断依据）\r\n\r\n### 蓝皮书不参与审计判断\r\n\r\n- 检查器**不得**从 `kw.get(\"blueprint\")` 拿数据做审计判断（已删除）\r\n- 审计判断**必须**用实时磁盘数据（`os.listdir` / `os.walk` / 读文件内容）\r\n- `audit_skill()` 已删除 `blueprint` 参数（v2.75.0），`_audit_with_blueprint()` 简化为废弃别名\r\n\r\n---\r\n\r\n## 2. 三个模式的完整流程\r\n\r\n### 2.1 创建（create）"},{"kind":"example","language":"text","snippet":"File v2.103.0:references/faq.md\n\n# 常见问题（FAQ）\r\n\r\n> 本文件收集 skill-standardization v2 使用过程中的常见疑问和解答。\r\n> 按「使用场景 → 技术细节 → 最佳实践」组织。\r\n\r\n---\r\n\r\n## 目录\r\n\r\n1. [基础概念](#基础概念)\r\n2. [create 模式](#create-模式)\r\n3. [update 模式](#update-模式)\r\n4. [refactor 模式](#refactor-模式)\r\n5. [审查与规范](#审查与规范)\r\n6. [渐进式 MD 体系](#渐进式-md-体系)\r\n7. [版本管理](#版本管理)\r\n8. [集成与扩展](#集成与扩展)\r\n\r\n---\r\n\r\n## 基础概念\r\n\r\n### Q1: 什么是 SKILL.md 标准化规范？\r\n\r\n**A:** 这是 skill-standardization 的核心规范——一套定义了标准 Skill 文件应如何编写的规则体系（v2+）。主要包含：\r\n\r\n- **Frontmatter 规范**：3 个必须字段（name/version/description）+ 7 个可选字段\r\n- **正文结构规范**：三层体系（must_have 必须章节 / whitelist 白名单章节 / nonstandard 非标章节）+ 渐进式索引表\r\n- **审查规则**：R-01~R-26 共 25 条自动检查规则（含安全审计、写作规范、章节顺位、格式合规等）\r\n\r\n该规范定义在 `scripts/spec/frontmatter.json`、`scripts/spec/body.json` 和 `scripts/spec/rules.json` 中，可直接查看 `scripts/spec/` 下的 JSON 文件。\r\n\r\n### Q2: skill-standardization 可以独立使用吗？\r\n\r\n**A:** 可以。skill-standardization 是一个独立审查工具，不依赖任何其他技能："}]}}