Skip to content
Back to search

claude-skills-mcp

SkillWorks with Claude CodeGitHub

MCP server for searching and retrieving Scientific Agent Skills using vector search

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/k-dense-ai--claude-skills-mcp/badge.svg)](/listings/k-dense-ai--claude-skills-mcp)

HTML

<a href="/listings/k-dense-ai--claude-skills-mcp"><img src="/api/listings/k-dense-ai--claude-skills-mcp/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.

For Cursor Users

Add through the [Cursor Directory](https://cursor.directory/mcp/claude-skills-mcp), or add to your Cursor config (`~/.cursor/mcp.json`):

{
  "mcpServers": {
    "claude-skills": {
      "command": "uvx",
      "args": ["claude-skills-mcp"]
    }
  }
}

The frontend starts instantly and displays tools, automatically downloading and starting the backend in the background (~60-120s due to RAG dependencies, one-time). Subsequent uses are instant.

Using uvx (Standalone)

Run the server with default configuration:

uvx claude-skills-mcp

This starts the lightweight frontend which auto-downloads the backend and loads 160+ skills from Anthropic's official skills repository and K-Dense's Scientific Agent Skills collection.

With Custom Configuration

# 1. Print the default configuration
uvx claude-skills-mcp --example-config > config.json

# 2. Edit config.json to your needs

# 3. Run with your custom configuration
uvx claude-skills-mcp --config config.json

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 · 159 lines

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

  1. # Claude Skills MCP Server
  2. > **This MCP server is no longer hosted or maintained.** Agent Skills have been natively adopted by all major AI platforms — Cursor, Windsurf, Claude Code, Copilot, and others now support skills out of the box. There is no longer a need for an MCP bridge to deliver skills to your coding assistant. The final release (v1.1.0) remains installable from PyPI. Thank you to everyone who used and contributed to this project!
  3. ---
  4. [![Tests](https://github.com/K-Dense-AI/claude-skills-mcp/actions/workflows/test.yml/badge.svg)](https://github.com/K-Dense-AI/claude-skills-mcp/actions/workflows/test.yml)
  5. [![Python 3.12 | 3.13](https://img.shields.io/badge/python-3.12%20%7C%203.13-blue.svg)](https://www.python.org/downloads/)
  6. [![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
  7. [![Code style: ruff](https://img.shields.io/badge/code%20style-ruff-000000.svg)](https://github.com/astral-sh/ruff)
  8. [![PyPI version](https://badge.fury.io/py/claude-skills-mcp.svg?icon=si%3Apython)](https://badge.fury.io/py/claude-skills-mcp)
  9. > **Use [Claude's powerful Skills system](https://www.anthropic.com/news/skills) with ANY AI model or coding assistant** - including Cursor, Codex, GPT-5, Gemini, and more. This MCP server brings Anthropic's Agent Skills framework to the entire AI ecosystem through the Model Context Protocol.
  10. A Model Context Protocol (MCP) server that provides intelligent search capabilities for discovering relevant Claude Agent Skills using vector embeddings and semantic similarity. This server implements the same progressive disclosure architecture that Anthropic describes in their [Agent Skills engineering blog](https://www.anthropic.com/engineering/equipping-agents-for-the-real-world-with-agent-skills), making specialized skills available to any MCP-compatible AI application.
  11. **An open-source project by [K-Dense](https://www.k-dense.ai)** - creators of autonomous AI scientists for scientific research.
  12. This MCP server enables any MCP-compatible AI assistant to intelligently search and retrieve skills from our curated [Scientific Agent Skills](https://github.com/K-Dense-AI/claude-scientific-skills) repository and other skill sources like the [Official Claude Skills](https://github.com/anthropics/skills).
  13. ## Demo
  14. ![Claude Skills MCP in Action](docs/demo.gif)
  15. *Semantic search and progressive loading of Claude Agent Skills in Cursor*
  16. ## Highlights
  17. - **Two-Package Architecture**: Lightweight frontend (~15 MB) starts instantly; backend (~250 MB) downloads in background
  18. - **No Cursor Timeout**: Frontend responds in <5 seconds, solving the timeout issue
  19. - **Semantic Search**: Vector embeddings for intelligent skill discovery
  20. - **Progressive Disclosure**: Multi-level skill loading (metadata → full content → files)
  21. - **Zero Configuration**: Works out of the box with curated skills
  22. - **Multi-Source**: Load from GitHub repositories and local directories
  23. - **Auto-Updating**: Checks skill sources hourly and re-indexes when they change
  24. - **Fast & Local**: No API keys needed, with automatic GitHub caching
  25. - **Configurable**: Customize sources, models, and content limits
  26. ## Quick Start
  27. ### For Cursor Users
  28. Add through the [Cursor Directory](https://cursor.directory/mcp/claude-skills-mcp), or add to your Cursor config (`~/.cursor/mcp.json`):
  29. ```json
  30. {
  31. "mcpServers": {
  32. "claude-skills": {
  33. "command": "uvx",
  34. "args": ["claude-skills-mcp"]
  35. }
  36. }
  37. }
  38. ```
  39. The frontend starts instantly and displays tools, automatically downloading and starting the backend in the background (~60-120s due to RAG dependencies, one-time). Subsequent uses are instant.
  40. ### Using uvx (Standalone)
  41. Run the server with default configuration:
  42. ```bash
  43. uvx claude-skills-mcp
  44. ```
  45. This starts the lightweight frontend which auto-downloads the backend and loads 160+ skills from Anthropic's official skills repository and K-Dense's Scientific Agent Skills collection.
  46. ### With Custom Configuration
  47. ```bash
  48. # 1. Print the default configuration
  49. uvx claude-skills-mcp --example-config > config.json
  50. # 2. Edit config.json to your needs
  51. # 3. Run with your custom configuration
  52. uvx claude-skills-mcp --config config.json
  53. ```
  54. ## Documentation
  55. - **[Getting Started](docs/getting-started.md)** - Installation, Cursor setup, CLI usage, and troubleshooting
  56. - **[Architecture Guide](docs/architecture.md)** - Two-package design, data flow, and components
  57. - **[API Documentation](docs/api.md)** - Tool parameters, examples, and best practices
  58. - **[Usage Examples](docs/usage.md)** - Advanced configuration, real-world use cases, and custom skill creation
  59. - **[Auto-Update Feature](docs/auto-update.md)** - Automatic hourly skill source refresh and change detection
  60. - **[Testing Guide](docs/testing.md)** - Complete testing instructions, CI/CD, and coverage analysis
  61. ## MCP Tools
  62. The server provides three tools for working with Claude Agent Skills:
  63. 1. **`find_helpful_skills`** - Semantic search for relevant skills based on task description
  64. 2. **`read_skill_document`** - Retrieve specific files (scripts, data, references) from skills
  65. 3. **`list_skills`** - View complete inventory of all loaded skills (for exploration/debugging)
  66. See [API Documentation](docs/api.md) for detailed parameters, examples, and best practices.
  67. ## Architecture
  68. The system uses a **two-package architecture** for optimal performance:
  69. - **Frontend** ([`claude-skills-mcp`](https://pypi.org/project/claude-skills-mcp/)): Lightweight proxy (~15 MB)
  70. - Starts instantly (<5 seconds) ✅ **No Cursor timeout!**
  71. - Auto-downloads backend on first use
  72. - MCP server (stdio) for Cursor
  73. - **Backend** ([`claude-skills-mcp-backend`](https://pypi.org/project/claude-skills-mcp-backend/)): Heavy server (~250 MB)
  74. - Vector search with PyTorch & sentence-transformers
  75. - MCP server (streamable HTTP)
  76. - Auto-installed by frontend OR deployable standalone
  77. **Benefits:**
  78. - ✅ Solves Cursor timeout issue (frontend starts instantly)
  79. - ✅ Same simple user experience (`uvx claude-skills-mcp`)
  80. - ✅ Backend downloads in background (doesn't block Cursor)
  81. - ✅ Can connect to remote hosted backend (no local install needed)
  82. See [Architecture Guide](docs/architecture.md) for detailed design and data flow.
  83. ## Skill Sources
  84. Load skills from **GitHub repositories** (direct skills or Claude Code plugins) or **local directories**.
  85. By default, loads from:
  86. - [Official Anthropic Skills](https://github.com/anthropics/skills) - diverse skills for documents, presentations, web artifacts, and more
  87. - [K-Dense Scientific Agent Skills](https://github.com/K-Dense-AI/claude-scientific-skills) - 148+ specialized skills for bioinformatics, cheminformatics, and scientific analysis
  88. - Local directory `~/.claude/skills` (if it exists)
  89. ## Contributing
  90. As this project is no longer maintained, new issues and pull requests may not be reviewed. If you'd like to build on it, please fork the repository. For reference, the original workflow was:
  91. 1. **Report issues**: [Open an issue](https://github.com/K-Dense-AI/claude-skills-mcp/issues) for bugs or feature requests
  92. 2. **Submit PRs**: Fork, create a feature branch, ensure tests pass (`uv run pytest tests/`), then submit
  93. 3. **Code style**: Run `uvx ruff check src/` before committing
  94. 4. **Add tests**: New features should include tests
  95. ### Development
  96. **Version Management**: This monorepo uses a centralized version system:
  97. - Edit the `VERSION` file at the repo root to bump the version
  98. - Run `python3 scripts/sync-version.py` to sync all references (or use `--check` to verify)
  99. - The `scripts/build-all.sh` script automatically syncs versions before building
  100. For questions, email [orion.li@k-dense.ai](mailto:orion.li@k-dense.ai)
  101. ## Learn More
  102. - [Agent Skills Documentation](https://docs.claude.com/en/docs/claude-code/skills) - Official Anthropic documentation on the Skills format
  103. - [Agent Skills Blog Post](https://www.anthropic.com/news/skills) - Announcement and overview
  104. - [Model Context Protocol](https://modelcontextprotocol.io/) - The protocol that makes cross-platform Skills possible
  105. - [Engineering Blog: Equipping Agents for the Real World](https://www.anthropic.com/engineering/equipping-agents-for-the-real-world-with-agent-skills) - Technical deep-dive on the Skills architecture
  106. ## License
  107. This project is licensed under the [Apache License 2.0](LICENSE).
  108. Copyright 2025-2026 K-Dense (https://www.k-dense.ai)

Ratings and comments

Sign in to rate this skill.

Sign in

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