mcp-server-demo
Exposes database and external API tools via the Model Context Protocol, enabling LLMs to query SQLite and call public APIs with validated inputs and error handling.
README
<p align="center"> <a href="https://www.linkedin.com/in/soriamaximilianorodrigo/" target="_blank" rel="noopener noreferrer"> <img width="100%" height="100%" src="docs/img/banner.gif" alt="mcp-server-demo"></a> </p>
<p align="center"> <a href="#"><img src="https://img.shields.io/badge/MCP-Model_Context_Protocol-14B8A6?logo=anthropic&logoColor=white&style=flat-square" alt="MCP"></a> <a href="#"><img src="https://img.shields.io/badge/Python_%2F_TypeScript-FastMCP-3776AB?style=flat-square" alt="SDK"></a> <a href="#"><img src="https://img.shields.io/badge/DB-SQLite-003B57?logo=sqlite&logoColor=white&style=flat-square" alt="SQLite"></a> <a href="#"><img src="https://img.shields.io/badge/Transporte-stdio_%7C_HTTP-22D3EE?style=flat-square" alt="Transporte"></a> </p>
<p align="center"> <a href="https://github.com/DietrichGebert/ponytail"><img src="https://img.shields.io/badge/built_with-ponytail-111111?style=flat-square" alt="ponytail"></a> <img src="https://img.shields.io/badge/layout-src%2Fpackage-14B8A6?style=flat-square" alt="src layout"> <img src="https://img.shields.io/badge/license-MIT-success?style=flat-square" alt="MIT"> </p>
<p align="center"> <img src="https://readme-typing-svg.demolab.com?font=Fira+Code&weight=600&size=22&pause=1000&color=1DE9B6¢er=true&vCenter=true&width=820&lines=Model+Context+Protocol%3A+tools+para+tu+LLM;SQLite+%2B+API+externa+expuestas+como+tools;stdio+local+o+HTTP+remoto" alt="typing SVG"> </p>
<hr/>
<h1 align="center">mcp-server-demo</h1>
<p align="center"> Servidor <b>MCP (Model Context Protocol)</b> que expone herramientas propias (consultas a una base SQLite y a una API externa) para que un LLM las use de forma estandarizada. </p>
Objetivo
Demostrar dominio del protocolo abierto que está estandarizando cómo los LLM acceden a herramientas y datos externos. En vez de acoplar tools a un framework concreto, se construye un servidor MCP reutilizable por cualquier cliente compatible (Claude Desktop, el sistema multi-agente de este portfolio, IDEs, etc.).
El servidor expone, como mínimo:
- Tools — acciones invocables: consultar una base de datos local (SQLite) y llamar a una API externa (ej. clima, tipo de cambio, o una API pública de datos).
- Resources — datos legibles por el modelo (ej. el esquema de la base, un catálogo).
- (Opcional) Prompts — plantillas de prompt reutilizables parametrizadas.
Stack tecnológico
- Opción A — Python: SDK oficial
mcp(conFastMCP) — recomendado por rapidez de desarrollo - Opción B — TypeScript:
@modelcontextprotocol/sdk— útil si se quiere mostrar versatilidad - Transporte:
stdiopara uso local con clientes de escritorio; Streamable HTTP para exposición remota - Base de datos: SQLite (con un dataset de ejemplo sembrado)
- API externa: una API pública sin fricción (ej. Open-Meteo para clima, o exchangerate.host)
- Validación: Pydantic (Python) o Zod (TS)
- Cliente de prueba: MCP Inspector (
npx @modelcontextprotocol/inspector) y/o Claude Desktop - Testing / calidad: pytest (o vitest), ruff/black (o eslint/prettier)
Estructura de carpetas (variante Python)
mcp-server-demo/
├── README.md
├── pyproject.toml
├── .env.example
├── data/
│ ├── seed.sql # Esquema + datos de ejemplo
│ └── demo.db # SQLite generada (git-ignored)
├── src/
│ ├── server.py # Entry point: instancia FastMCP y registra todo
│ ├── config.py
│ ├── tools/
│ │ ├── db_tools.py # query_customers, get_order_by_id, etc.
│ │ └── api_tools.py # get_weather, get_exchange_rate, etc.
│ ├── resources/
│ │ └── schema.py # Expone el esquema de la DB como resource
│ └── db/
│ └── connection.py # Conexión y helpers (queries parametrizadas)
├── scripts/
│ └── seed_db.py # Crea demo.db desde seed.sql
├── examples/
│ └── claude_desktop_config.json # Config de ejemplo para registrar el server
└── tests/
├── test_db_tools.py
└── test_api_tools.py
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.