Agent Skills API
A multi-agent system for code rules management using MCP, supporting search, generation, and architectural validation through specialized agents.
README
Agent Skills API - CodeMentor MCP
🤖 Multi-Agent System con Arquitectura Hexagonal + CQRS
Sistema de agentes especializados para búsqueda y gestión de reglas de código usando el protocolo MCP (Model Context Protocol).
🚀 Quick Start
1. Instalar dependencias
pnpm install
2. Configurar variables de entorno
cp .env.example .env
3. Build
pnpm run build
4. Iniciar servidor
pnpm run start:dev
5. Verificar
curl http://localhost:8004/health
📚 Documentación
| Documento | Descripción |
|---|---|
| README.md | Guía principal y quick start |
| ARCHITECTURE.md | Arquitectura v3: monorepo, gRPC, agentes |
| MCP-CONFIG.md | Configuración MCP por agente (Qwen, Cursor, Claude) |
| MCP-QWEN-CONFIG.md | Configuración específica para Qwen + reglas |
| IMPLEMENTATION_STATUS.md | Estado de implementación y métricas |
🤖 Agentes Disponibles
| Agente | Responsabilidad |
|---|---|
| RouterAgent | Orquestador principal, detecta intención |
| SearchAgent | Búsqueda BM25 de reglas |
| IdentityAgent | Gestiona identidad MCP y prefijos |
| RulesAgent | Listado y gestión de reglas |
| CodeAgent | Generación de código |
| ArchitectureAgent | Validación arquitectónica |
| AnalysisAgent | Análisis de código |
| MetricsAgent | Métricas y tracking |
🤖 Agentes CLI Compatibles
| Agente | Configuración | Estado |
|---|---|---|
| Qwen Code | qwen-mcp-config.json |
✅ Configurado |
| OpenCode | .opencode.json |
✅ Configurado |
| Gemini CLI | gemini-mcp-config.json |
✅ Configurado |
| Cursor | .cursor/settings.json |
📝 Ver doc/MCP-CONFIG.md |
| Claude Code | ~/.claude/mcp.json |
📝 Ver doc/MCP-CONFIG.md |
🛠️ Scripts Disponibles
# Desarrollo
pnpm run start:dev
# Producción
pnpm run start:prod
# Build
pnpm run build
# Tests
pnpm run test
# Lint
pnpm run lint
# Gemini Setup
pnpm run agent:setup-gemini # Setup Gemini MCP config
pnpm run agent:setup-gemini:force # Force overwrite existing config
📡 Endpoints
| Método | Endpoint | Descripción |
|---|---|---|
GET |
/health |
Health check |
GET |
/rules |
Listar reglas |
GET |
/rules/search?q=xxx |
Buscar reglas |
GET |
/mcp/sse |
MCP SSE endpoint |
GET |
/api |
Swagger UI |
🔧 Configuración Gemini
Para configurar Gemini CLI con Agent Skills API:
# Copiar configuración
cp gemini-mcp-config.json ~/.gemini/mcp.json
# Editar la ruta del servidor MCP si es necesario
nano ~/.gemini/mcp.json
Ver .gemini/INSTRUCTIONS.md para más detalles.
🔧 Configuración Qwen
Para priorizar MCP en Qwen, usa la configuración en doc/MCP-QWEN-CONFIG.md:
{
"mcp": {
"enabled": true,
"autoActivateSkills": true,
"preferMcpOverInternalTools": true
},
"mcpServers": {
"agent-skills-api": {
"url": "http://localhost:8004/mcp/sse",
"trust": true,
"priority": "high"
}
}
}
📊 Métricas
| Componente | Valor |
|---|---|
| Agentes | 8 especializados |
| Reglas | 20+ reglas de código |
| Tests | 17 tests passing |
| Arquitectura | Hexagonal + CQRS |
🎯 Reglas Aplicadas
Todos los agentes siguen estas reglas:
- ✅ Comentarios en inglés
- ✅ Logs en inglés
- ✅ Respuestas amigables (no técnicas)
- ✅ Clean Architecture
- ✅ CQRS pattern
- ✅ Repository pattern
📝 License
MIT
Para más detalles, visita la carpeta doc/
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.