MCP Servers

Hubify can integrate with AI tools via the Model Context Protocol (MCP).

Overview

MCP provides a standardized way to give AI models access to external tools and context. Hubify offers an MCP server for deep integration.

Hubify MCP Server

Install and configure the Hubify MCP server:

npm install -g @hubify/mcp

Configuration

Add to your MCP client configuration:

{
  "mcpServers": {
    "hubify": {
      "command": "hubify-mcp",
      "args": ["serve"],
      "env": {
        "HUBIFY_API_KEY": "your-api-key"
      }
    }
  }
}

Available Tools

The Hubify MCP server exposes these tools:

skill_search

Search for skills in the registry:

{
  "name": "skill_search",
  "arguments": {
    "query": "react hooks",
    "limit": 10
  }
}

skill_install

Install a skill:

{
  "name": "skill_install",
  "arguments": {
    "name": "typescript-patterns",
    "global": false
  }
}

skill_info

Get skill details:

{
  "name": "skill_info",
  "arguments": {
    "name": "typescript-patterns"
  }
}

learning_report

Report execution results:

{
  "name": "learning_report",
  "arguments": {
    "skill": "typescript-patterns",
    "result": "success",
    "improvement": "Add async error handling pattern"
  }
}

Resources

The server exposes installed skills as MCP resources:

hubify://skills/typescript-patterns
hubify://skills/react-best-practices

AI models can read these directly for context.

Prompts

Pre-configured prompts for common tasks:

apply_skill

{
  "name": "apply_skill",
  "arguments": {
    "skill": "typescript-patterns",
    "context": "refactoring a module"
  }
}

report_result

{
  "name": "report_result",
  "arguments": {
    "skill": "typescript-patterns",
    "success": true
  }
}

Client Configuration

Claude Desktop

{
  "mcpServers": {
    "hubify": {
      "command": "npx",
      "args": ["@hubify/mcp", "serve"]
    }
  }
}

Windsurf

{
  "mcpServers": {
    "hubify": {
      "command": "hubify-mcp",
      "args": ["serve"]
    }
  }
}

Building Custom MCP Servers

Use the Hubify SDK to build custom MCP integrations:

import { HubifyClient } from "@hubify/sdk";
import { Server } from "@modelcontextprotocol/sdk/server";

const hubify = new HubifyClient();
const server = new Server({ name: "my-hubify-server" });

server.setRequestHandler("tools/call", async (request) => {
  if (request.params.name === "search_skills") {
    const results = await hubify.skills.search(request.params.arguments.query);
    return { content: [{ type: "text", text: JSON.stringify(results) }] };
  }
});

Learn More

MCP Specification

Learn more about the Model Context Protocol