NaliiaBotMCP/AGENTS.md

TrytonMCP Agent Specification

Desarrollo

Configuración inicial

# Instalar dependencias con uv
uv sync

# Copiar configuración de ejemplo
cp .env.example .env

# Instalar dependencias de desarrollo y test
uv pip install -e ".[test]"

Comandos frecuentes

# Ejecutar servidor MCP (stdio)
uv run fastmcp run src/tryton_mcp/server.py

# Ejecutar tests
python -m pytest tests/ -v

# Verificar código
ruff check src/ tests/
ruff format src/ tests/

Docker

# Construir imagen
docker compose build

# Iniciar contenedor
docker compose up -d

# Ver logs
docker compose logs -f

# Detener
docker compose down

Agregar nueva herramienta

from tryton_mcp.services.base import TrytonService

class YourService(TrytonService):
    name = "your.model.method"

    def execute(self, params: dict) -> List[dict]:
        # Implementar lógica
        pass

Overview

TrytonMCP es un servidor MCP (Model Context Protocol) que permite a los LLMs interactuar con Tryton ERP a través de llamadas RPC. Envuelve la librería sabatron-tryton_rpc_client y expone los modelos de Tryton como herramientas invocables.

Project Structure

TrytonMCP/
├── src/
│   └── tryton_mcp/
│       ├── __init__.py
│       ├── config.py          # Settings singleton with TrytonSettings
│       ├── server.py          # Main MCP server with tool definitions
│       └── services/
│           ├── base.py        # TrytonService base class
│           ├── party.py       # Customer operations
│           ├── product.py     # Products/Services
│           ├── schedule.py    # Appointments
│           └── service_center.py  # Service centers
├── tests/                     # Test suite
│   ├── conftest.py
│   ├── test_customer.py
│   └── test_server.py
├── test_rpc.py                # Standalone RPC test script
├── Dockerfile                # Docker container definition
├── docker-compose.yaml        # Docker Compose orchestration
├── .dockerignore              # Docker build exclusions
├── pyproject.toml            # Project dependencies
└── .env.example              # Environment variables template

Architecture

Dependencies

• fastmcp[tasks]>=3.1.0 - MCP server framework
• sabatron-tryton-rpc-client>=7.4.0 - Tryton RPC client

Core Components

1. Settings (config.py) - Singleton pattern for configuration management

   • Uses environment variables with defaults
   • Provides to_dict() and get_client() methods
   • Validates required fields in __post_init__

2. Tools (server.py) - Functions decorated with @mcp.tool():

   • create_schedule - Creates appointments in Naliia module
   • find_service_centers - Searches service centers
   • find_products_and_services - Searches salable products

Adding New Tools

To add a new Tryton model as an MCP tool:

@mcp.tool()
def your_tool_name(param1: int, param2: str) -> List[dict]:
    client = settings.get_client()
    client.connect()
    
    name = "model.your.model.method"
    args = [params, ...]
    response = client.call(name, args)
    
    return response

Running the Server

# Run MCP server (stdio mode)
uv run fastmcp run src/tryton_mcp/server.py

Configuration

Credentials are loaded from environment variables (with sensible defaults):

export TRYTON_HOSTNAME=dev.naliia.co
export TRYTON_DATABASE=capacitacion
export TRYTON_USERNAME=admin
export TRYTON_PASSWORD=admin
export TRYTON_PORT=8000

Or copy .env.example to .env and adjust values:

cp .env.example .env

Note: .env is gitignored to protect credentials.

Testing RPC Calls

Use test_rpc.py to test Tryton RPC calls independently:

python test_rpc.py