- Explore MCP Servers
- MCP-Guide
Mcp Guide
What is Mcp Guide
MCP-Guide is a repository that demonstrates the Model Context Protocol (MCP) concept by integrating Large Language Models (LLMs) such as Claude/Anthropic with external tools using the MCP protocol, employing a microservices architecture.
Use cases
Use cases include developing intelligent chatbots that can interact with databases, creating dynamic workflows using LLMs, and building advanced AI applications that leverage external tools through the MCP protocol.
How to use
To use MCP-Guide, clone the repository, configure the MCP server and chatbot by setting environment variables, install dependencies, and run the services locally or via Docker.
Key features
Key features include a FastAPI-based MCP server for CRUD operations on PostgreSQL, a web-based chatbot interface using Streamlit that integrates with the Claude LLM, and a standardized architecture for intelligent agents to utilize external tools.
Where to use
MCP-Guide can be used in fields such as artificial intelligence, chatbot development, and any application requiring integration of LLMs with external data sources or services.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Overview
What is Mcp Guide
MCP-Guide is a repository that demonstrates the Model Context Protocol (MCP) concept by integrating Large Language Models (LLMs) such as Claude/Anthropic with external tools using the MCP protocol, employing a microservices architecture.
Use cases
Use cases include developing intelligent chatbots that can interact with databases, creating dynamic workflows using LLMs, and building advanced AI applications that leverage external tools through the MCP protocol.
How to use
To use MCP-Guide, clone the repository, configure the MCP server and chatbot by setting environment variables, install dependencies, and run the services locally or via Docker.
Key features
Key features include a FastAPI-based MCP server for CRUD operations on PostgreSQL, a web-based chatbot interface using Streamlit that integrates with the Claude LLM, and a standardized architecture for intelligent agents to utilize external tools.
Where to use
MCP-Guide can be used in fields such as artificial intelligence, chatbot development, and any application requiring integration of LLMs with external data sources or services.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Content
MCP Guide
Este repositório demonstra o conceito de MCP (Model Context Protocol) integrando LLMs (Claude/Anthropic) com ferramentas externas via protocolo MCP, utilizando uma arquitetura de microsserviços.
Visão Geral
O projeto é composto por dois principais microsserviços:
- ms_mcp: Servidor MCP implementado com FastAPI, responsável por expor ferramentas de CRUD de clientes em PostgreSQL via protocolo MCP.
- ms_chat: Chatbot com interface web (Streamlit) que integra o LLM Claude (Anthropic) e se conecta ao servidor MCP para executar comandos e fluxos dinâmicos.
A arquitetura permite que agentes inteligentes (LLMs) utilizem ferramentas externas de forma padronizada, facilitando a criação de aplicações avançadas de IA.
Estrutura do Projeto
MCP-Guide/ │ ├── ms_mcp/ # Servidor MCP (FastAPI, PostgreSQL) │ ├── mcp_server.py │ ├── api/ │ ├── infra/ │ ├── requirements.txt │ └── README.MD │ ├── ms_chat/ # Chatbot com Claude LLM + MCP Client (Streamlit) │ ├── app.py │ ├── client_server.py │ ├── requirements.txt │ └── README.MD │ └── README.md # Este arquivo
Como Executar
1. Clonar o repositório
git clone https://github.com/Leonardojdss/MCP-Guide.git
cd MCP-Guide
2. Configurar e rodar o servidor MCP
Veja instruções detalhadas em ms_mcp/README.MD.
Resumo:
- Configure o banco Azure PostgreSQL ou local e variáveis de ambiente no
.env. - Instale dependências:
pip install -r ms_mcp/requirements.txt - Rode localmente:
uvicorn mcp_server:app --host 0.0.0.0 --port 8000 - Ou via Docker.
3. Rodar o chatbot com Claude
Veja instruções detalhadas em ms_chat/README.MD.
Resumo:
- Configure sua chave da Anthropic no
.env. - Instale dependências:
pip install -r ms_chat/requirements.txt - Rode localmente:
streamlit run ms_chat/app.py - Ou via Docker.
Exemplo de Uso
- Inicie o servidor MCP.
- Inicie o chatbot.
- No chatbot, conecte-se ao endpoint do servidor MCP (ex:
http://localhost:8000/mcp). - Interaja com o assistente, que poderá executar comandos CRUD em clientes via MCP.
Dependências Principais
- Python 3.10+
- FastAPI, FastMCP, fastapi-mcp
- Anthropic (Claude)
- Streamlit
- PostgreSQL
Documentação dos Microsserviços
- ms_mcp/README.MD: Documentação do servidor MCP.
- ms_chat/README.MD: Documentação do chatbot e integração com Claude.
Scripts Úteis
Exemplo de tabela no PostgreSQL
CREATE TABLE IF NOT EXISTS customer (
id SERIAL PRIMARY KEY,
name VARCHAR(100) NOT NULL,
email VARCHAR(100) UNIQUE,
phone VARCHAR(20),
address TEXT,
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);
Este projeto é experimental e serve para explorar o conceito de MCP, integração de LLMs e ferramentas externas.
Dev Tools Supporting MCP
The following are the main code editors that support the Model Context Protocol. Click the link to visit the official website for more information.










