mcp-skills
Integrates with the skills.sh ecosystem to allow AI coding agents to discover, install, and manage reusable instruction sets. It enables autonomous agents to extend their capabilities with structured skill discovery and full lifecycle management through the Model Context Protocol.
README
mcp-skills
What is mcp-skills?
mcp-skills is an MCP server that integrates with the open skills.sh ecosystem. It enables coding agents and LLMs to discover, recommend, and install reusable skills automatically.
Why this matters
Modern AI coding agents can write code, but they can’t easily extend themselves.
When an agent lacks a capability (e.g. React testing, refactoring patterns, architectural scaffolding), it usually depends on:
- Manual CLI execution
- Human intervention
- Prompt hacks
mcp-skills bridges that gap.
It gives AI agents structured access to the open skills.sh ecosystem through Model Context Protocol (MCP), enabling:
- Skills discovery
- Context-aware skill recommendation and installation
- Full lifecycle management (install, update, uninstall)
What is skills.sh?
skills.sh is The Open Agent Skills Ecosystem by Vercel. It lets you search, install, and manage skills (reusable instruction sets) for popular coding agents like Copilot, Cursor, Claude Code, Codex, OpenCode, and many more.
Example use cases:
-
"Find a skill to generate React components" → Agent searches and installs it
-
"Install the TypeScript refactoring skill" → Agent adds it to your environment
-
Agent detects missing capability → Autonomously searches and installs the right skill
-
"Analyze this monorepo and enforce a consistent testing strategy across all packages" → Agent detects Jest/Vitest usage, installs the appropriate testing skill, and refactors configs automatically
-
"Migrate this legacy Express API to a typed architecture" → Agent selects a TypeScript migration skill, installs it, and applies structured refactors across the codebase
mcp-skills vs skills CLI
-
CLI: Token-efficient and fast for direct execution.
-
MCP: Better for structured reasoning, introspection, and autonomous agent workflows.
Getting started
Quick Installation (Production)
Install the MCP server from npm. No need to clone the repository.
Standard config works in most MCP clients:
{ "mcpServers": { "mcp-skills": { "command": "npx", "args": [ "-y", "@koderspa/mcp-skills@latest" ] } } }
<details> <summary>Amp</summary>
Add via the Amp VS Code extension settings screen or by updating your settings.json file:
"amp.mcpServers": { "mcp-skills": { "command": "npx", "args": [ "@koderspa/mcp-skills@latest" ] } }Amp CLI Setup:
Add via the
amp mcp addcommand belowamp mcp add mcp-skills -- npx @koderspa/mcp-skills@latest</details>
<details> <summary>Claude Code</summary>
Use the Claude Code CLI to add the mcp-skills MCP server:
claude mcp add mcp-skills npx @koderspa/mcp-skills@latest</details>
<details> <summary>Claude Desktop</summary>
Follow the MCP install guide, use the standard config above.
</details>
<details> <summary>Cline</summary>
Follow the instruction in the section Configuring MCP Servers
Example: Local Setup
Add the following to your
cline_mcp_settings.jsonfile:{ "mcpServers": { "mcp-skills": { "type": "stdio", "command": "npx", "timeout": 30, "args": [ "@koderspa/mcp-skills@latest" ], "disabled": false } } }</details>
<details> <summary>Codex</summary>
Use the Codex CLI to add the mcp-skills MCP server:
codex mcp add mcp-skills npx "@koderspa/mcp-skills@latest"Alternatively, create or edit the configuration file
~/.codex/config.tomland add:[mcp_servers.mcp-skills] command = "npx" args = ["@koderspa/mcp-skills@latest"]For more information, see the Codex MCP documentation.
</details>
<details> <summary>Copilot</summary>
Use the Copilot CLI to interactively add the mcp-skills MCP server:
/mcp addAlternatively, create or edit the configuration file
~/.copilot/mcp-config.jsonand add:{ "mcpServers": { "mcp-skills": { "type": "local", "command": "npx", "tools": [ "*" ], "args": [ "@koderspa/mcp-skills@latest" ] } } }For more information, see the Copilot CLI documentation.
</details> <details> <summary>VS Code</summary>
Follow the MCP install guide, use the standard config above. You can also install the mcp-skills MCP server using the VS Code CLI:
# For VS Code code --add-mcp '{"name":"mcp-skills","command":"npx","args":["@koderspa/mcp-skills@latest"]}'After installation, the mcp-skills MCP server will be available for use with your GitHub Copilot agent in VS Code. </details> <details> <summary>Cursor</summary>
Click the button to install:
<img src="https://cursor.com/deeplink/mcp-install-dark.svg" alt="Install in Cursor">
Or install manually:
Go to
Cursor Settings->MCP->Add new MCP Server. Name to your liking, usecommandtype with the commandnpx @koderspa/mcp-skills@latest. You can also verify config or add command like arguments via clickingEdit.</details>
<details> <summary>Factory</summary>
Use the Factory CLI to add the mcp-skills MCP server:
droid mcp add mcp-skills "npx @koderspa/mcp-skills@latest"Alternatively, type
/mcpwithin Factory droid to open an interactive UI for managing MCP servers.For more information, see the Factory MCP documentation.
</details>
<details> <summary>Gemini CLI</summary>
Follow the MCP install guide, use the standard config above.
</details>
<details> <summary>Goose</summary>
Click the button to install:
Or install manually:
Go to
Advanced settings->Extensions->Add custom extension. Name to your liking, use typeSTDIO, and set thecommandtonpx @koderspa/mcp-skills@latest. Click "Add Extension". </details><details> <summary>Kiro</summary>
Follow the MCP Servers documentation. For example in
.kiro/settings/mcp.json:{ "mcpServers": { "mcp-skills": { "command": "npx", "args": [ "@koderspa/mcp-skills@latest" ] } } }</details>
<details> <summary>LM Studio</summary>
Click the button to install:
Or install manually:
Go to
Programin the right sidebar ->Install->Edit mcp.json. Use the standard config above. </details><details> <summary>opencode</summary>
Follow the MCP Servers documentation. For example in
~/.config/opencode/opencode.json:{ "$schema": "https://opencode.ai/config.json", "mcp": { "mcp-skills": { "type": "local", "command": [ "npx", "@koderspa/mcp-skills@latest" ], "enabled": true } } }</details>
<details> <summary>Qodo Gen</summary>
Open Qodo Gen chat panel in VSCode or IntelliJ → Connect more tools → + Add new MCP → Paste the standard config above.
Click <code>Save</code>. </details>
<details> <summary>Warp</summary>
Go to
Settings->AI->Manage MCP Servers->+ Addto add an MCP Server. Use the standard config above.Alternatively, use the slash command
/add-mcpin the Warp prompt and paste the standard config from above:{ "mcpServers": { "mcp-skills": { "command": "npx", "args": [ "@koderspa/mcp-skills@latest" ] } } }</details>
<details> <summary>Windsurf</summary>
Follow Windsurf MCP documentation. Use the standard config above.
</details>
Overview
AI Agent ↓ MCP Client ↓ mcp-skills (this server) ↓ skills.sh CLI ↓ Installed Skills
Tools
<details> <summary><strong>findSkills — Search for skills by keyword</summary>
Read-only: Yes Input: Query string Output: List of matching skills with metadata
Uses the skills.sh CLI to search the open skills ecosystem. </details>
<details> <summary><strong>installSkill — Install a skill by ID</summary>
Read-only: No (modifies system)
Input: Skill ID (format: <code>owner/repo@skill</code>)
Output: Installation success/failure message
Auto-detects: Installs missing skills CLI automatically
Installs a skill from skills.sh or GitHub by its ID. </details>
<details> <summary><strong>uninstallSkill — Uninstall a skill by ID</summary>
Read-only: No (modifies system)
Input: Skill ID (format: <code>owner/repo@skill</code>)
Output: Uninstallation success/failure message
Removes a previously installed skill by its ID. </details>
<details> <summary><strong>updateSkill — Update a skill by ID</summary>
Read-only: No (modifies system)
Input: Skill ID (format: <code>owner/repo@skill</code>)
Output: Update success/failure message
Updates an installed skill to its latest version. </details>
<details> <summary><strong>selectSkill — Recommend the most relevant skill for a task</summary>
Read-only: Yes
Input: Task description
Output: Best matching skill with details
Recommends the most relevant skill for a specific task using the skills.sh ecosystem. </details>
<details> <summary><strong>analyzeProjectStack — Analyze project stack and recommend skills</summary>
Read-only: Yes
Input: None (analyzes <code>package.json</code> in the current directory)
Output: Detected stack and recommended skills
Analyzes your project's <code>package.json</code> to detect frameworks/libraries and recommends relevant skills. </details>
Resources
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
- Fork the repository
- Create your feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add some amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
Support
- Issues: GitHub Issues
- Email: yusterfrancisco@koderspa.com
License
Apache-2.0 License - see LICENSE file for details.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.