agentleFS
Sign inStart free
echoleesongecholeesong/claude-skills-plugin/skills/skill-manager/SKILL.md

skill-manager

Skill4 starsChanged 5 months agoMITView on GitHub

Skill 生命周期管理器。用于批量扫描 Skills 目录、检查 GitHub 更新、执行版本升级、管理 Skills 库存、管理经验层（promote/pull）。当用户需要"检查更新"、"列出 Skills"、"删除 Skill"、"管理 Skills"、"提升经验"、"拉取经验"时使用此工具。

52 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use Skill Evolution Manager, github-to-skills and 7 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • Arthur-devroleSkill

    Skill Evolution Manager

    专门用于在对话结束时，根据用户反馈和对话内容总结优化并迭代现有 Skills 的核心工具。它通过吸取对话中的“精华”（如成功的解决方案、失败的教训、特定的代码规范）来持续演进 Skills 库。

    Also in 5 projects that use this

  • Arthur-devroleSkill

    github-to-skills

    Automated factory for converting GitHub repositories into specialized AI skills. Use this skill when the user provides a GitHub URL and wants to "package", "wrap", or "create a skill" from it. It automatically fetches repository details, latest commit hashes, and generates a standardized skill structure with enhanced metadata suitable for lifecycle management.

    Also in 5 projects that use this

  • NVIDIASkill

    review-code

    Review a code repository for concrete issues in correctness, robustness, security, maintainability, tests, and integration. Use for repository-wide or focused code review where practical engineering judgment and strict scope control matter.

    Also in 3 projects that use this

  • OpenAISkill

    brainstorming

    You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

    Also in 4 projects that use this

  • Microsoft AzureSkill

    humanizer

    Tone enforcement patterns for external-facing community responses

    Also in 3 projects that use this

  • AnthropicSkill

    pdf

    Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Claude needs to fill in a PDF form or programmatically process, generate, or analyze PDF documents at scale.

    Also in 3 projects that use this

The file

Create a free account to mount
Files
SKILL.md

Skill Manager

管理 Skills 的完整生命周期：扫描、检查更新、升级、删除、经验层管理。

核心功能

  1. 扫描审计: 扫描 Skills 目录，解析所有 SKILL.md 的 frontmatter 元数据
  2. 版本检查: 并发查询 GitHub（通过 git ls-remote）比较本地和远程 hash
  3. 状态报告: 生成状态报告，标识 outdated/current/unmanaged/error
  4. 更新工作流: 引导 Agent 执行 Skill 升级
  5. 库存管理: 列出、删除 Skills
  6. 经验层管理: promote/pull 命令在层级间同步经验

启动流程（重要）

每次调用 skill-manager 时，必须首先执行以下流程确定 Skills 目录：

步骤 1: 检查配置

运行 python scripts/config.py get 检查是否已配置 Skills 目录。

步骤 2: 根据配置状态处理

情况 A - 已配置:

  1. 向用户确认："将管理 <skills_dir> 中的 Skills，确认继续？"
  2. 用户确认 → 继续执行操作
  3. 用户拒绝 → 询问新路径，运行 python scripts/config.py set <new_path>

情况 B - 未配置:

  1. 运行 python scripts/config.py detect 自动检测
  2. 检测成功 → 向用户确认检测到的路径，确认后继续
  3. 检测失败 → 询问用户 Skills 目录路径，然后运行 python scripts/config.py set <path>

示例对话

Agent: 检查 skill-manager 配置...
Agent: 已配置 Skills 目录: /Users/xxx/.claude/skills
Agent: 将管理该目录中的 Skills，确认继续？
User: 确认
Agent: [执行用户请求的操作]
Agent: 检查 skill-manager 配置...
Agent: 未找到配置，正在自动检测...
Agent: 检测到 Skills 目录: /Users/xxx/.claude/skills
Agent: 是否使用该目录？
User: 是
Agent: 已保存配置。[执行用户请求的操作]

注意: 不要探索当前工作目录来寻找 Skills，Skills 目录是用户专门存放 Skills 的位置，与当前项目无关。

使用场景

检查更新

触发方式:

  • /skill-manager check
  • "检查我的 Skills 是否有更新"
  • "扫描 Skills 更新"

工作流:

  1. Agent 运行 scripts/scan_and_check.py <skills_dir>
  2. 脚本并发检查所有带有 source_url 的 Skills
  3. 输出 JSON 格式的状态报告
  4. Agent 向用户呈现结果

示例输出:

Skills 状态摘要
========================================
总计: 12 个 Skills
  - 最新: 8
  - 需更新: 2
  - 非托管: 1
  - 错误: 1

需要更新的 Skills:
  - yt-dlp: New commits available
    本地: abc123...
    远程: def456...

列出 Skills

触发方式:

  • /skill-manager list
  • "列出我的 Skills"
  • "显示所有 Skills"

工作流:

  1. Agent 运行 scripts/list_skills.py <skills_dir>
  2. 输出表格或 JSON 格式的 Skills 列表

示例输出:

Name                   | Type     | Ver      | Description
-----------------------|----------|----------|------------------------------------------
md-to-pptx             | Standard | 1.0.0    | Markdown 转 PowerPoint 工具...
n8n-workflow-patterns  | Standard | 1.0.0    | n8n 工作流模式专家...
yt-dlp                 | GitHub   | 0.1.0    | YouTube 视频下载工具封装...
-----------------------|----------|----------|------------------------------------------
Total: 12 skills

更新 Skill

触发方式:

  • "更新 [Skill 名称]"（在检查更新后）
  • /skill-manager update <skill_name>

工作流:

  1. Agent 从远程仓库获取新的 README
  2. 对比新旧 README，识别变化（新功能、废弃参数等）
  3. 重写 SKILL.md 反映新功能
  4. 更新 source_hash 字段
  5. 可选：更新 wrapper.py
  6. 运行 skill-evolution 的 align_all.py 恢复用户经验

删除 Skill

触发方式:

  • /skill-manager delete <skill_name>
  • "删除 Skill: <skill_name>"

工作流:

  1. Agent 运行 scripts/delete_skill.py <skill_name> <skills_dir>
  2. 脚本显示 Skill 信息并请求确认
  3. 可选：备份到 .skill-backup 目录
  4. 执行删除

经验层管理

查看经验状态

触发方式:

  • /skill-manager evo-status <skill_name>
  • "查看 xxx 的经验状态"

工作流:

python scripts/evo_manager.py evo-status <skill_name> [--project <path>]

示例输出:

Skill: n8n-code-javascript
Project: /Users/xxx/my-project

Layers:
  [✗] Upstream
      Path: ~/.claude/skills-repo/skills/n8n-code-javascript/evolution.json
  [✓] Global
      Path: ~/.claude/evolutions/n8n-code-javascript/evolution.json
      Items: 5 total
        - preferences: 2
        - fixes: 2
        - contexts: 1
  [✓] Project
      Path: /Users/xxx/my-project/.claude/evolutions/n8n-code-javascript.json
      Items: 3 total
        - preferences: 1
        - fixes: 2

Merged Total: 8 items

提升经验到全局层

触发方式:

  • /skill-manager promote <skill_name>
  • "把这个项目的经验提升到全局"

工作流:

python scripts/evo_manager.py promote <skill_name> [--project <path>] [--fields f1,f2]

将当前项目的经验复制到全局层，供所有项目共享。

从全局层拉取经验

触发方式:

  • /skill-manager pull <skill_name>
  • "把全局经验拉到这个项目"

工作流:

python scripts/evo_manager.py pull <skill_name> [--project <path>]

将全局层的经验复制到当前项目，用于新项目初始化。

同步所有经验状态

触发方式:

  • /skill-manager sync-evolutions
  • "查看所有 Skills 的经验状态"

工作流:

python scripts/evo_manager.py sync [--skills-dir <path>] [--project <path>]

示例输出:

============================================================
Skills Evolution Status Summary
============================================================
Skills Directory: ~/.claude/skills-repo/skills
Project Path: /Users/xxx/my-project
Total Skills: 14

Summary:
  - Has upstream evolution: 2
  - Has global evolution: 5
  - Has project evolution: 3
  - No evolution data: 8

Skills Detail:
------------------------------------------------------------
Name                            Up   Glb   Prj  Total
------------------------------------------------------------
md-to-pptx                       -     ✓     -      3
n8n-code-javascript              -     ✓     ✓      8
n8n-code-python                  -     -     -      0
...
------------------------------------------------------------

脚本说明

脚本功能参数
config.py配置管理get|set <path>|detect|clear
scan_and_check.py扫描并检查更新<skills_dir> [--json|--summary|--table]
list_skills.py列出所有 Skills<skills_dir> [--json] [--verbose]
delete_skill.py删除 Skill<skill_name> <skills_dir> [--backup] [--force]
evo_manager.py经验层管理<command> <skill_name> [options]

evo_manager.py 命令

命令功能
evo-status <skill>查看经验分布状态
promote <skill>项目经验 → 全局层
pull <skill>全局经验 → 项目层
sync查看所有 skills 经验状态

元数据依赖

此管理器依赖以下元数据字段：

字段用途
source_url远程仓库 URL（用于检查更新）
source_hash本地记录的 commit hash
version语义化版本号
evolution_enabled是否启用经验进化

兼容性: 同时支持旧格式的 github_url 和 github_hash 字段。

状态说明

状态含义
current本地 hash 与远程相同，无需更新
outdated远程有新的 commits
unmanaged没有配置 source_url，无法检查更新
error无法访问远程仓库
unknown本地没有记录 hash

最佳实践

  1. 定期检查: 建议每周运行一次 /skill-manager check
  2. 备份删除: 删除重要 Skill 前使用 --backup 选项
  3. 更新后对齐: 更新 Skill 后运行 skill-evolution 的 align_all.py 恢复经验
  4. 版本控制: 使用 Git 管理 Skills 目录，便于回滚
  5. 经验管理:
    • 新项目开始时运行 pull 获取全局经验
    • 项目结束时运行 promote 提升有价值的经验
    • 定期运行 sync 查看经验分布

与其他 Skill 的协作

  • skill-factory: 创建的 Skills 自动包含生命周期管理所需的元数据
  • skill-evolution: 更新后调用 align_all.py 恢复用户经验；共享 layered_merge.py
  • skill-creator: 遵循相同的元数据规范

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free