ContextVM MCP Bridge
Enables LLMs to interact with ContextVM services (Event Sourcing + FSM + DDD) by translating Model Context Protocol requests to DVM commands over Nostr, allowing AI agents to manage domain entities and workflows.
README
ContextVM MCP Bridge
Expone servicios ContextVM (Event Sourcing + FSM + DDD) a LLMs mediante Model Context Protocol sobre Nostr.
🎯 ¿Qué hace este Bridge?
Este bridge actúa como traductor bidireccional entre:
- MCP (Model Context Protocol): Protocolo estándar para que LLMs interactúen con servicios externos
- ContextVM: Nuestra arquitectura de Event Sourcing + FSM + DDD
LLMs (Claude, GPT, etc.)
↓ MCP (kind 25910)
MCP Bridge (Traductor)
↓ DVM (kind 5051-6054)
ContextVM Services (sin cambios)
✨ Características
- ✅ No invasivo: Los ContextVMs existentes no se modifican
- ✅ Nostr como BUS: Un solo relay para MCP y DVM
- ✅ Bidireccional: Traduce peticiones y respuestas
- ✅ Configurable: Mapeos externos en JSON
- ✅ Observable: Logs detallados y métricas
- ✅ Production-ready: Docker, health checks, error handling
🚀 Quick Start
1. Instalación
# Clonar repositorio
git clone https://github.com/your-org/contextvm-mcp-bridge.git
cd contextvm-mcp-bridge
# Instalar dependencias
npm install
# Configurar
cp .env.example .env
# Editar .env con tu configuración
2. Generar Keypair
# Generar nueva keypair para el bridge
npx tsx scripts/generate-keypair.ts
# Copiar el private key a .env
# BRIDGE_PRIVATE_KEY=<hex-key>
3. Configurar Contextos
Editar config/contexts.json con los ContextVMs disponibles:
{
"contexts": [
{
"id": "cbz_tesoreria_pagos",
"name": "CBZ Tesorería - Pagos Ejecutados",
"namespace": "cbz-tesoreria-pagos-ejecutados",
"jobRequestKind": 5053,
"jobResultKind": 6053,
"rootEntity": {
"type": "pago-tesoreria",
"initialState": "planificado",
"schema": { ... },
"transitions": [ ... ]
}
}
]
}
4. Ejecutar
# Modo desarrollo
npm run dev
# Modo producción
npm run build
npm start
# Con Docker
docker-compose up -d
📋 Uso para LLMs
Conectar desde Claude/GPT
import { NostrMCPProxy } from '@contextvm/sdk/proxy';
// Crear proxy MCP
const proxy = new NostrMCPProxy({
privateKey: "llm-private-key-hex",
relays: ["wss://relay.controller-ai.com"],
serverPubkey: "bridge-public-key-hex"
});
await proxy.start();
// Listar herramientas disponibles
const tools = await proxy.request({
method: "tools/list"
});
// Llamar herramienta
const result = await proxy.request({
method: "tools/call",
params: {
name: "cbz_tesoreria_pagos_create",
arguments: {
importe: 1500.00,
concepto: "Pago nómina",
beneficiario: "Juan Pérez"
}
}
});
🏗️ Arquitectura
src/
├── core/
│ ├── bridge.ts # Clase principal del bridge
│ ├── translator.ts # MCP ↔ DVM translator
│ └── routing.ts # Routing logic
├── mcp/
│ ├── mcp-listener.ts # Escucha kind 25910
│ ├── mcp-publisher.ts # Publica kind 25910
│ └── mcp-types.ts # Tipos MCP
├── dvm/
│ ├── dvm-listener.ts # Escucha kind 5051-6054
│ ├── dvm-publisher.ts # Publica kind 5051-6054
│ └── dvm-types.ts # Tipos DVM
├── mapping/
│ ├── capabilities.ts # Mapeo de capabilities
│ ├── schemas.ts # Schemas de validación
│ └── contexts.ts # Contextos disponibles
├── config/
│ └── index.ts # Configuración
└── index.ts # Entry point
🔧 Configuración Avanzada
Mapeo de Contextos
Ver docs/context-mapping.md para detalles sobre cómo mapear ContextVMs a MCP tools.
Queries sin Eventos
Para consultas de solo lectura, el bridge puede consultar directamente la API/PostgreSQL sin generar eventos DVM:
// En config/contexts.json
{
"queries": {
"enabled": true,
"apiEndpoint": "https://api.cbz-tesoreria.controller-ai.com",
"cacheEnabled": true,
"cacheTTL": 300
}
}
📊 Monitoring
Health Check
curl http://localhost:4000/health
# Response:
{
"status": "healthy",
"uptime": 3600,
"relay": "connected",
"contexts": 4,
"requests": {
"total": 1234,
"success": 1200,
"errors": 34
}
}
Métricas
El bridge expone métricas en formato Prometheus:
curl http://localhost:4000/metrics
🧪 Testing
# Tests unitarios
npm test
# Tests con watch
npm run test:watch
# Tests de integración (requiere relay activo)
npm run test:integration
🐳 Docker
Build
docker build -t contextvm-mcp-bridge:latest .
Run
docker run -d \
--name mcp-bridge \
-p 4000:4000 \
-e BRIDGE_PRIVATE_KEY=your-key \
-e RELAY_URL=ws://relay:8080 \
contextvm-mcp-bridge:latest
Docker Compose
docker-compose up -d
📖 Documentación
🤝 Contribuir
- Fork el repositorio
- Crear feature branch (
git checkout -b feature/amazing-feature) - Commit cambios (
git commit -m 'Add amazing feature') - Push al branch (
git push origin feature/amazing-feature) - Abrir Pull Request
📄 Licencia
MIT License - ver LICENSE
🔗 Links
- ContextVM Docs: https://contextvm.org
- MCP Protocol: https://modelcontextprotocol.io
- Nostr Protocol: https://nostr.com
- GitHub: https://github.com/your-org/contextvm-mcp-bridge
Última actualización: 23 de diciembre de 2025
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.