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 agentmods add commands/cn-big-cabbage/github-skill-distiller/generate-skillgit clone --depth 1 https://github.com/CN-big-cabbage/github-skill-distillerWrote 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/commands/cn-big-cabbage/github-skill-distiller/generate-skill)<a href="https://agentmods.dev/commands/cn-big-cabbage/github-skill-distiller/generate-skill"><img src="https://agentmods.dev/badge/commands/cn-big-cabbage/github-skill-distiller/generate-skill.svg" alt="Measured on agentmods" 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 | $0.00025 | $0.03124 |
| Opus 5 | $0.00013 | $0.01562 |
| Sonnet 5 | $0.00005 | $0.00625 |
| Haiku 4.5 | $0.00003 | $0.00312 |
Grade A, and why
generate-skill 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 3d 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 — 274 lines — stays where its author put it; the contents beside it link to each section on GitHub.
自动技能生成
你将根据用户提供的 GitHub URL，自动完成技能创建的全流程。
输入
用户提供一个 GitHub 仓库 URL，例如：
/generate-skill https://github.com/soimort/you-get
全流程
完整流程：Phase 1 → Phase 1.5 → Phase 2 → Phase 2.5 → Phase 3a（评分）→ Phase 3b（修复）→ Phase 3c（决策） → Phase 4（验证）→ Phase 5（用户确认）→ Phase 6（发布）
Phase 1: 项目分析
- 运行分析脚本：
./scripts/analyze-repo.sh <github-url> /tmp/skill-gen-<repo-name>
- 读取生成的
/tmp/skill-gen-<repo-name>/project-profile.json - 读取克隆仓库的 README 文件（路径在 profile 的 clone_dir + readme_file 中）
- 如果 has_docs 为 true，浏览 docs/ 目录获取更多上下文
Phase 1.5: 类型适配策略
根据 project-profile.json 中的 project_type 字段选择对应的生成策略：
CLI 工具（project_type: "cli"）
- SKILL.md 重点: 命令速查表、常用参数组合、输出格式说明
- guides 重点: 按使用场景组织（如"下载视频"、"批量操作"），每个场景给出完整命令
- troubleshooting 重点: 参数错误、权限问题、环境变量配置
库（project_type: "library"）
- SKILL.md 重点: API 参考、代码示例、返回值说明
- guides 重点: 集成到项目中的步骤、常用模式和最佳实践
- troubleshooting 重点: 导入错误、版本冲突、类型不匹配
框架（project_type: "framework"）
- SKILL.md 重点: 架构概览、核心概念、扩展点
- guides 重点: 项目脚手架搭建、路由/中间件/插件配置
- troubleshooting 重点: 配置错误、插件兼容性、启动失败
Web 应用（project_type: "webapp"）
- SKILL.md 重点: 功能清单、部署选项、架构图
- guides 重点: 本地开发环境搭建、部署流程
- troubleshooting 重点: 端口冲突、依赖缺失、环境变量
DevOps 工具（project_type: "devops"）
- SKILL.md 重点: 架构图、组件关系、支持的平台
- guides 重点: 环境准备、集群部署、配置定制
- troubleshooting 重点: 网络问题、权限配置、版本兼容
Phase 2: 生成技能内容
在 /tmp/skill-gen-<repo-name>/skill/ 目录下生成以下文件：
2.1 生成 SKILL.md
参考 cases/you-get/SKILL.md 的质量标准，生成包含以下内容的 SKILL.md：
- frontmatter: name, description, version(0.1.0), metadata(openclaw requires, emoji, homepage)
- 技能概述: 说明技能帮助谁完成什么，列出 3-5 个核心场景
- 使用流程: AI 引导的步骤（4-6 步）
- 关键章节导航: 链接到 guides/ 和 troubleshooting.md
- AI 助手能力: 6-8 条 AI 可执行的具体能力
- 核心功能: 功能清单（用 ✅ 标记）
- 快速示例: 3-5 个常用命令示例
- 安装要求: 列出依赖
- 项目链接: GitHub、官网、文档
2.2 生成 guides/
根据项目复杂度生成 1-3 个 guide 文件：
guides/01-installation.md— 安装指南（必须）guides/02-quickstart.md— 快速开始（必须）guides/03-advanced-usage.md— 高级用法（复杂项目）
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.
- 3d ago First seen · 274 lines · 25 tokens per session scan A f013b8b2fa7d
generate-skill is a command published in the GitHub repository CN-big-cabbage/github-skill-distiller (4 stars, last pushed 2mo ago), licensed MIT. It adds 25 tokens to every session and 3,124 once invoked, about $0.0001 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 commands, from other repositories
capture-this-chat
Turn the current chat session into a reusable spell.
build-spell
Start building a new spell from a task you do repeatedly.
share-spell
Export a spell for sharing with others.
list-spells
Browse all available spells (yours plus bundled).
refine-spell
Improve an existing spell based on real usage.
cast-spell
Run a spell from your library.