mcp-registry-interface

MikeyBeez/mcp-registry-interface

3.3

If you are the rightful owner of mcp-registry-interface and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

The MCP Registry Interface is a server that facilitates the discovery, browsing, and management of Model Context Protocol (MCP) servers from the official registry.

Tools
5
Resources
0
Prompts
0

mcp-registry-interface

๐Ÿ” MCP server for interfacing with the official Model Context Protocol registry

Enables Claude and other MCP clients to dynamically discover, browse, search, and manage MCP servers directly from GitHub. Fetches real-time data from the official MCP servers repository.

๐Ÿš€ Features

  • ๐Ÿ” Real-time Search - Find MCP servers from live GitHub data
  • ๐Ÿ“ฆ Server Details - Get comprehensive information including installation guides
  • ๐Ÿ“‚ Categories - Browse servers by functionality (filesystem, database, etc.)
  • ๐Ÿ”„ Auto-refresh - Cached data with configurable refresh intervals
  • ๐Ÿ”’ Rate Limit Aware - Supports GitHub tokens for higher API limits
  • โŒ No Mock Data - Always provides real data or fails transparently

๐Ÿ“‹ Available Tools

ToolDescription
registry_search_serversSearch MCP servers by name, description, or tags
registry_get_server_detailsGet detailed server information and installation guides
registry_list_categoriesList all server categories with descriptions
registry_refresh_dataForce refresh GitHub data (bypasses cache)

๐Ÿ› ๏ธ Installation

NPM (Recommended)

npx mcp-registry-interface

Manual Installation

git clone <repository-url>
cd mcp-registry-interface
npm install
npm start

โš™๏ธ Configuration

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "registry-interface": {
      "command": "npx",
      "args": ["mcp-registry-interface"],
      "env": {
        "GITHUB_TOKEN": "your-github-token-optional"
      }
    }
  }
}

Environment Variables

VariableDescriptionRequired
GITHUB_TOKENGitHub personal access token for higher rate limitsNo (60 requests/hour without)

Without a GitHub token, you get 60 API calls per hour. With a token, you get 5,000 calls per hour.

๐ŸŽฏ Example Usage

User: "Find MCP servers for working with databases"
Claude: Uses registry_search_servers with category="database"

User: "Show me details about the filesystem server"
Claude: Uses registry_get_server_details with serverId="mcp-filesystem"

User: "What categories of MCP servers are available?"
Claude: Uses registry_list_categories

๐Ÿ”’ Security & Reliability

  • โœ… Real Data Only - No mock/fake data that could mislead developers
  • โœ… Transparent Failures - Clear error messages when GitHub API is unavailable
  • โœ… Rate Limit Handling - Proper GitHub API rate limit management
  • โœ… Input Validation - Secure parameter checking and sanitization
  • โœ… No Hardcoded Secrets - All tokens from environment variables

๐Ÿงช Development & Testing

The server fetches live data from GitHub's API:

# Basic usage (60 requests/hour)
node src/index.js

# With GitHub token (5000 requests/hour) 
GITHUB_TOKEN=your_token_here node src/index.js

# Test the server
node static-test.js

๐Ÿ›ก๏ธ Error Handling

When GitHub API is unavailable, the server:

  • โŒ Does NOT provide fake/mock data
  • โœ… Returns clear error messages
  • โœ… Explains how to resolve the issue
  • โœ… Maintains cache for recent successful requests

๐Ÿค Contributing

This tool helps accelerate the MCP ecosystem by making server discovery seamless with real, up-to-date information.

  1. Fork the repository
  2. Create a feature branch
  3. Make your changes
  4. Test with live GitHub data
  5. Submit a pull request

๐Ÿ“š Data Source

  • Primary Source: modelcontextprotocol/servers README
  • API: GitHub Contents API
  • Update Frequency: 5-minute cache with manual refresh capability
  • Parsing: Extracts official reference servers and community servers

๐Ÿ“„ License

MIT License - see LICENSE file for details.


๐ŸŽฏ Philosophy: This tool believes in transparency - real data or honest failure, never misleading mock data.