Skip to content

openmodelsrun/mcp

BranchesTags

Repository files navigation

OpenModels MCP Server Registry

A community-maintained registry of MCP (Model Context Protocol) servers — structured, searchable, and integrated with the OpenModels ecosystem.

Stats

Metric Count
Servers 32
Categories 11
Transport types 3 (stdio, sse, http-streaming)

What is this?

This registry catalogs MCP servers that extend AI agents with tools, resources, and prompts. Each server entry includes:

  • Description — what the server does and what capabilities it exposes
  • Tools — available tool definitions with input schemas
  • Resources & Prompts — exposed resources and prompt templates
  • Install config — how to install and run the server
  • Metadata — category, tags, transport protocols, author, license

Directory Structure

mcp/
├── servers/                 # One YAML file per MCP server
│   ├── _template.yaml      # Template for contributors
│   ├── arxiv.yaml
│   ├── aws-docs.yaml
│   ├── brave-search.yaml
│   ├── cloudflare.yaml
│   ├── context7.yaml
│   ├── docker.yaml
│   ├── exa.yaml
│   ├── fetch.yaml
│   ├── firecrawl.yaml
│   ├── filesystem.yaml
│   ├── git.yaml
│   ├── github.yaml
│   ├── grafana.yaml
│   ├── linear.yaml
│   ├── memory.yaml
│   ├── mongodb.yaml
│   ├── neon.yaml
│   ├── notion.yaml
│   ├── perplexity.yaml
│   ├── playwright.yaml
│   ├── postgres.yaml
│   ├── puppeteer.yaml
│   ├── qdrant.yaml
│   ├── redis.yaml
│   ├── sentry.yaml
│   ├── sequential-thinking.yaml
│   ├── slack.yaml
│   ├── sqlite.yaml
│   ├── stripe.yaml
│   ├── supabase.yaml
│   ├── tavily.yaml
│   └── time.yaml
├── categories/
│   └── categories.yaml     # Category definitions
├── schemas/
│   └── server.schema.json  # JSON Schema (Draft 7) for validation
├── validate.py             # Validation script
├── requirements.txt        # Python dependencies (PyYAML, jsonschema)
├── CONTRIBUTING.md          # Contribution guide
├── CHANGELOG.md             # Version history
├── LICENSE                  # MIT
├── README.md                # This file
└── VERSION                  # Current version (0.3.0)

Quick Start

Browse servers

Visit openmodels.run/mcp to search, filter, and compare MCP servers with a web UI.

Or explore the YAML files directly in the servers/ directory.

Install a server

Each server detail page at openmodels.run/mcp/{server-id} includes an Install MCP server button that copies the install command to your clipboard and provides IDE-specific configuration snippets.

Add a new server

  1. Fork this repository
  2. Copy servers/_template.yaml to servers/{your-server-id}.yaml
  3. Fill in your server details (see CONTRIBUTING.md)
  4. Validate locally: python validate.py
  5. Submit a pull request

Server Format

Each server is a YAML file validated against a JSON Schema:

id: github
name: GitHub MCP Server
description: >
  Provides access to the GitHub API through the Model Context Protocol...
author:
  name: GitHub
  github: github
repository: https://github.com/github/github-mcp-server
transport:
  - stdio
  - http-streaming
category: Development
tags:
  - github
  - git
  - version-control
tools:
  - name: create_issue
    description: Create a new issue in a GitHub repository.
    input_schema:
      type: object
      properties:
        owner:
          type: string
        repo:
          type: string
        title:
          type: string
      required: [owner, repo, title]
install:
  npm: "@github/github-mcp-server"
  command: docker
  args: ["run", "-i", "--rm", "-e", "GITHUB_PERSONAL_ACCESS_TOKEN", "ghcr.io/github/github-mcp-server"]
env_vars:
  - name: GITHUB_PERSONAL_ACCESS_TOKEN
    description: GitHub personal access token for API authentication
    required: true
license: MIT
version: "1.0.5"
stars: 30200
created_at: "2024-11-25T00:00:00.000Z"
updated_at: "2026-05-21T00:00:00.000Z"

Servers

Server Category Author Transport
ArXiv Research Joe Blazick stdio, http-streaming
AWS Docs Development AWS Labs stdio
Brave Search Research Brave stdio
Cloudflare Cloud Cloudflare stdio
Context7 Development Upstash stdio, http-streaming
Docker DevOps Docker stdio
Exa Research Exa Labs stdio
Fetch Development Anthropic stdio
Firecrawl Research Mendable stdio
Filesystem Filesystem Anthropic stdio
Git Development Anthropic stdio
GitHub Development GitHub stdio, http-streaming
Grafana DevOps Grafana Labs stdio, sse
Linear Productivity Linear stdio
Memory AI Anthropic stdio
MongoDB Database MongoDB stdio, http-streaming
Neon Database Neon stdio
Notion Productivity Notion stdio
Perplexity Research Perplexity AI stdio, http-streaming
Playwright Browser Automation Microsoft stdio
PostgreSQL Database Anthropic stdio
Puppeteer Browser Automation Anthropic stdio
Qdrant Database Qdrant stdio, sse, http-streaming
Redis Database Anthropic stdio
Sentry DevOps Sentry stdio
Sequential Thinking AI Anthropic stdio
Slack Communication Anthropic stdio
SQLite Database Anthropic stdio
Stripe Development Stripe stdio
Supabase Database Supabase stdio
Tavily Research Tavily stdio
Time Productivity Anthropic stdio

Categories

Category Description
Development Code generation, testing, debugging, and software development tools
Productivity Task management, note-taking, workflow automation
Database Database querying, schema management, and data manipulation
DevOps CI/CD, infrastructure management, monitoring
AI Machine learning, model management, and AI utilities
Research Web search, knowledge retrieval, and information gathering
Browser Automation Web scraping, browser control, and automated testing
Communication Messaging, email, notifications, and collaboration
Filesystem File operations, directory management, and local storage
Cloud Cloud provider APIs, serverless functions, and cloud storage
Security Authentication, encryption, and vulnerability scanning

Validation

All server definitions are validated against JSON Schema on every PR:

pip install -r requirements.txt
python validate.py

Validation checks:

  • YAML syntax
  • Schema conformance (required fields, types, enums, patterns)
  • No duplicate IDs across all server definitions
  • Filename matches the id field

CI Integration

A GitHub Actions workflow runs validation automatically on every pull request that modifies files in servers/, schemas/, or validate.py. PRs that fail validation cannot be merged.

Contributing

We welcome contributions! See CONTRIBUTING.md for the full guide.

Quick steps:

  1. Fork this repository
  2. Copy servers/_template.yaml and fill in your server details
  3. Run validation: python validate.py
  4. Submit a pull request

Integration with OpenModels

MCP servers are ingested into the OpenModels platform and available via:

  • Web UI — browse, filter, search, and compare at openmodels.run/mcp
  • API — query programmatically via /api/v1/mcp/servers
  • IDE configs — generate connection snippets for Claude Code, Cursor, Windsurf, Gemini CLI, Kiro, and VS Code
  • Install button — one-click copy of install commands on each server detail page

License

MIT

About

Open-source registry of MCP servers – structured metadata, tools, resources and install configs for the Model Context Protocol ecosystem.

www.openmodels.run

Topics

open-source mcp llm-tools model-context-protocol mcp-server

Resources

Readme

License

MIT license

Contributing

Contributing
Activity
Custom properties

Stars

0 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

 
 
 

Contributors

Languages