agentes-mcp-server
Unified MCP server for the AmorSaúde ecosystem, combining SQL query, semantic search, medical record similarity, audio transcription, and LLM-as-Judge evaluation tools for multiple healthcare agents.
README
Servidor MCP Centralizado — AmorSaúde
Servidor MCP (Model Context Protocol) centralizado que unifica as ferramentas de consulta SQL (Athena), busca semântica (Pinecone RAG), busca de prontuários similares, transcrição de áudio (Whisper) e avaliação de qualidade (LLM-as-Judge) para todos os agentes do ecossistema AmorSaúde.
Agentes Suportados
| Agente | Athena | RAG | Busca Semântica | Transcrição | Evaluator |
|---|---|---|---|---|---|
| Cora | ✅ (Cardiologia/HAS) | HAS | ❌ | ❌ | ✅ Unificado |
| Amorzito | ✅ (Qualidade/IQRC) | CFM, Regras, RDC, HAS | ❌ | ✅ | ✅ Unificado |
| Iris | ✅ (Catarata) | Catarata, Vocabulário | ✅ | ❌ | ✅ Unificado |
| Auxiliar Médico | ❌ | CFM, Regras, RDC | ❌ | ✅ | ✅ Transcrição |
Primitivas MCP
🔧 Tools (5)
query_athena_tool(sql, agent_id)— Consulta SQL com validação por agentesearch_rag_tool(query, agent_id, namespace_key, k)— Busca semântica no Pineconesearch_similar_records_tool(query, agent_id, top_k)— Busca de prontuários similarestranscribe_audio_tool(file_path, agent_id)— Transcrição via Whisperevaluate_response_tool(agent_id, ...)— Avaliação LLM-as-Judge
📄 Resources (6)
agent://registry/list— Lista todos os agentesagent://{agent_id}/config— Configuração do agenteagent://{agent_id}/sql-rules— Regras SQL específicasagent://{agent_id}/rag-namespaces— Namespaces RAG disponíveisagent://{agent_id}/schema— Schema do banco de dadosagent://{agent_id}/evaluator-criteria— Critérios de avaliação
💬 Prompts (3)
setup-agent(agent_id, data_hoje, data_ontem)— System prompt completo do agentebuild-sql-expert-prompt(agent_id)— Prompt do sub-agente SQLbuild-evaluator-prompt(agent_id)— Prompt do avaliador LLM-as-Judge
Setup
# 1. Instalar dependências
uv sync
# 2. Configurar variáveis de ambiente
cp .env.example .env
# Edite o .env com os valores reais
# 3. Rodar o servidor (SSE)
uv run server.py
# 4. Testar com o MCP Inspector
npx @modelcontextprotocol/inspector uv run server.py
Estrutura
central-mcp-server/
├── server.py # Ponto de entrada FastMCP (SSE)
├── config/
│ ├── settings.py # Variáveis de ambiente (Pydantic Settings)
│ └── agents.py # Registry: AGENT_CONFIGS, PERSONAS, EVALUATOR_CONFIGS
├── tools/
│ ├── athena.py # query_athena (validação + execução)
│ ├── rag.py # search_rag (Pinecone unificado)
│ ├── semantic_search.py # search_similar_records (embedding direto)
│ ├── transcription.py # transcribe_audio (Whisper)
│ └── evaluator.py # evaluate_response (LLM-as-Judge)
├── resources/
│ └── agent_resources.py # Resources dinâmicos por agente
├── pyproject.toml
├── .env.example
└── README.md
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.