Skip to content
Back to search

skills-management

A CLI tool to manage and synchronize AI coding agent skills

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/nnnggel--skills-management/badge.svg)](/listings/nnnggel--skills-management)

HTML

<a href="/listings/nnnggel--skills-management"><img src="/api/listings/nnnggel--skills-management/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

This tool can be installed directly from npm:

npm install -g skills-management

Or install from source:

# Clone the repository
git clone https://github.com/nnnggel/skills-management.git
cd skills-management
npm install
npm run build
npm link

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 156 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # Skills Management CLI (skm)
  2. [简体中文](./README.zh-CN.md) | English
  3. `skm` is a powerful CLI tool designed to manage and synchronize "skills" (prompt libraries, instruction sets, or capability modules) across various AI coding agents and projects. It serves as a central hub to add skills from GitHub or local directories and selectively link them into your local AI project configurations.
  4. It supports multiple AI environments including **OpenCode**, **Cursor**, **Gemini**, **Antigravity**, **Claude**, and **GitHub** projects.
  5. ---
  6. ## 🚀 Features
  7. - **Global Skill Repository**: Centralized management of your AI skills.
  8. - **GitHub Integration**: Add skills directly from GitHub repositories or specific subdirectories (sparse checkout).
  9. - **Local Skill Support**: Add skills from local directories for development and debugging.
  10. - **Version Control**: Check for updates and synchronize changes from remote repositories.
  11. - **Project Detection**: Automatically detects the AI project type in your current directory.
  12. - **Symbolic Linking**: Efficiently links skills to projects without duplication, keeping them in sync.
  13. - **Project Isolation**: Manage different sets of skills for different projects.
  14. ![Screenshot](repo.png)
  15. ## 💡 Why skm?
  16. - **Centralized Efficiency**: Skills are cloned only once (`~/.skills-management/repo`) and shared across infinite projects via symbolic links. This saves significant disk space and ensures all your projects use the same curated version of a skill.
  17. - **Precision (Sparse Checkout)**: Only download what you need. If a GitHub repository contains hundreds of skills but you only want one, `skm` uses Git's sparse-checkout to download only that specific subdirectory. No more cloning massive repositories for a single prompt file.
  18. - **Automatic Environment Awareness**: You don't need to know where Cursor, Claude, or OpenCode stores their skills. `skm` automatically detects your project environment and handles the directory structures for you.
  19. - **Clean & Non-Intrusive**: Since skills are symlinked, your project folder stays clean. No extra `.git` folders or heavy files are added to your project's version control.
  20. - **Self-healing**: Proactively detects and helps you clean up broken symbolic links if a global skill is deleted.
  21. - **Version Sync**: Check for remote updates to keep your local "AI brain" sharp.
  22. ## 📦 Installation
  23. This tool can be installed directly from npm:
  24. ```bash
  25. npm install -g skills-management
  26. ```
  27. Or install from source:
  28. ```bash
  29. # Clone the repository
  30. git clone https://github.com/nnnggel/skills-management.git
  31. cd skills-management
  32. npm install
  33. npm run build
  34. npm link
  35. ```
  36. ## 📖 Usage
  37. Run the tool using `skm`:
  38. ```bash
  39. skm
  40. ```
  41. ### 1. Global Repository Management (`repo`)
  42. Select **"1. repo"** from the main menu to manage your global collection of skills.
  43. #### Add Skill
  44. **GitHub Type**:
  45. - Full repositories: `https://github.com/user/repo` (example: https://github.com/blader/humanizer)
  46. - Specific subfolders: `https://github.com/user/repo/tree/main/path/to/skill` (example: https://github.com/anthropics/skills/tree/main/skills/pdf)
  47. **Local Type**:
  48. - Local directories: `/local/path/to/skill` (example: `~/Desktop/myskill`)
  49. #### Other Operations
  50. - **Update Skills**: Checks for newer commits on the remote GitHub repository and updates your local copy (GitHub type only).
  51. - **Delete Skills**: Remove skills from your global repository.
  52. ### 2. Project Skill Management (`list`)
  53. Navigate to your AI project directory and run `skm`. The tool will detect the project type (e.g., `.opencode`, `.cursor`).
  54. Select **"2. list(type)"** to manage skills for the current project.
  55. - **Link/Unlink**: You will see a list of available global skills.
  56. - **Checkbox Interface**: Use `Space` to select or deselect skills, and `Enter` to confirm.
  57. - **Symlinks**: Selected skills are symlinked contents from your global repo into your project's skill directory (e.g., `.opencode/skills/`).
  58. ### Supported Project Structures
  59. `skm` automatically detects and installs skills into these directories:
  60. | AI Type | Skills Installation Path |
  61. |---------|---------------------------|
  62. | **Antigravity** | `.gemini/antigravity/global_skills/skills` or `.agent/skills` |
  63. | **GitHub** | `.copilot/skills` or `.github/skills` |
  64. | **Cursor** | `.cursor/skills` |
  65. | **Claude** | `.claude/skills` |
  66. | **OpenCode** | `.opencode/skills` |
  67. ---
  68. ## 🏗️ Directory Structure & Internals
  69. `skm` stores its global data and configuration in your home directory at `~/.skills-management`. This architecture ensures that skills are downloaded once and shared across multiple projects.
  70. ### Data Layout:
  71. - **`config.json`**: Global configuration file. It stores system-level settings and metadata.
  72. - **`repo/`**: The heart of the management system.
  73. - **`skills.json`**: The skill registry. It tracks all added skills, their unique IDs, current commit hashes (for version control), and original paths.
  74. - **`[type]__[name]/`**: Local skill repository storage. `skm` flattens the ID (replacing `/` with `__`) to create safe directory names. For example, `github:user/repo/path` becomes `github__user__repo__path`, and `local:myskill` becomes `local__myskill`.
  75. - **Note**: `skm` handles complex paths like `github:user/repo/path/to/skill` by creating a unique folder like `github__user__repo__path__to__skill`.
  76. ### Cross-Platform Compatibility (Windows)
  77. `skm` is fully compatible with Windows.
  78. - On **macOS/Linux**, it uses standard symbolic links.
  79. - On **Windows**, it automatically uses **Directory Junctions** (a type of symbolic link for folders) to ensure capability without requiring Administrator privileges or Developer Mode.
  80. ### Custom AI Tools Configuration
  81. You can customize the AI tools detection by adding `aiTools` to your `~/.skills-management/config.json`:
  82. ```json
  83. {
  84. "system": "darwin",
  85. "aiTools": [
  86. { "type": "cursor", "skillDirs": [".cursor/skills"] },
  87. { "type": "custom-ai", "skillDirs": [".custom-ai/skills", ".custom-ai/prompts"] }
  88. ]
  89. }
  90. ```
  91. Each entry requires:
  92. - **`type`**: A unique identifier for the AI tool (used in menu display)
  93. - **`skillDirs`**: An array of possible skill directory paths (relative to project root)
  94. > **Note**: When `aiTools` is configured, it completely replaces the default configuration. If you want to keep some defaults, include them in your custom configuration.
  95. ---
  96. ## 📝 Changelog
  97. ### v1.0.5
  98. - Display version number in main menu
  99. - Show non-skm managed skills when managing project skills
  100. - Support custom AI tools configuration via `config.json`
  101. ### v1.0.4
  102. - Adjusted AI tool detection directories to match official configurations
  103. ### v1.0.3
  104. - Local Skill Support: Add skills directly from local directories
  105. - Menu Navigation Optimization: Unified numeric menu style, added return options, improved terminal compatibility
  106. ## 📄 License
  107. ISC

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.