MCP App Directory
jit9499/mcpappdirectory
Descubra e compare os melhores servidores MCP para seus agentes IA
Ver no GitHubDiretório curado de servidores MCP (Model Context Protocol) com busca, comparação e ranking de qualidade. Contém mais de 1.100 servidores catalogados com algoritmo de scoring, analytics do ecossistema e submissões comunitárias.
Instalação
npx convez-ai jit9499/mcpappdirectory1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
Casos de uso
- Encontrar servidores MCP específicos por categoria e capacidade
- Comparar múltiplos servidores lado-a-lado antes de integrar
- Acompanhar tendências e top performers do ecossistema MCP
- Submeter novos servidores para catalogação comunitária
Veredito

Ferramenta essencial para quem trabalha com MCP servers. O diretório resolve um problema real de discovery no ecossistema, oferecendo busca, ranking transparente e comparação. Implementação técnica sólida com pipeline de enriquecimento de dados e scoring versionado.
README
A tradução para o português deste README ainda está na fila.
🔌 MCP App Directory
The curated directory of MCP (Model Context Protocol) servers — discover, compare, and choose the best MCP servers for your AI agents.
🌐 Live at mcpappdirectory.com
✨ Features
- 🔍 Search & Browse — Find MCP servers by category, capability, and use case
- 🏆 Leaderboard — Top-ranked servers scored by quality, reliability, and community trust
- ⚖️ Compare — Side-by-side comparison of MCP servers
- 📊 Dashboard — Analytics and insights about the MCP ecosystem
- 📝 Submit — Community-driven submissions of new MCP servers
- 📧 Newsletter — Stay updated on new MCP servers and ecosystem news
🏗️ Architecture
mcpappdirectory/
├── server.py # Main backend server (API + routing)
├── index.html # Directory listing page
├── leaderboard.html # Top-ranked MCP servers
├── compare.html # Side-by-side comparison tool
├── dashboard.html # Analytics dashboard
├── submit.html # Submit new MCP server
├── scoring.html # Scoring methodology
├── about.html / how-it-works.html
├── listings.json # MCP server data (1,100+ entries)
├── submissions.json # Community submissions
├── subscribers.json # Newsletter subscribers
├── score_v3.py # Scoring algorithm (v3)
├── rescore.py # Batch rescoring
├── rescore_github.py # GitHub-based rescoring
├── classify.py # Server classification
├── enrich_listings.py # Data enrichment pipeline
├── validate_*.py # URL & data validation
├── audit.py # Site auditing
└── health_check.py # Health monitoring
🚀 Quick Start
Prerequisites
- Python 3.10+
- pip
Setup
# Clone the repo
git clone https://github.com/jit9499/mcpappdirectory.git
cd mcpappdirectory
# Install dependencies
pip install -r requirements.txt # if available
# Run the server
python server.py
📈 Data Pipeline
The directory uses a multi-stage data pipeline:
- Classify (
classify.py) — Categorize MCP servers by type and capability - Enrich (
enrich_listings.py) — Pull additional metadata (GitHub stars, descriptions, etc.) - Score (
score_v3.py) — Rate servers on quality, reliability, and community metrics - Validate (
validate_*.py) — Verify URLs and data integrity - Display — Serve via the web frontend
🤝 Contributing
We welcome contributions! Here's how:
- Submit an MCP Server — Use the submission form on the website
- Report Issues — Open a GitHub Issue
- Code Contributions — Fork, create a branch, submit a PR
📊 Stats
- 1,100+ MCP servers listed
- Multi-version scoring algorithm
- Community-driven submissions
- SEO-optimized with sitemap & IndexNow
📜 License
This project is licensed under the MIT License — see the LICENSE file for details.
🔗 Related Projects
- awesome-mcp-servers — A curated list of MCP servers
- Model Context Protocol — The official MCP specification
Built with ❤️ by @jit9499