wyoloservice2_mcp
Allows AI agents to inspect and control YOLO training studies in a Train Service 2 / NeuralForgeAI cluster, including launching, canceling, and monitoring training jobs.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@wyoloservice2_mcpcheck the cluster status"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
wyoloservice2_mcp
Servidor de Model Context Protocol (MCP) para el clúster Train Service 2 / NeuralForgeAI. Permite a los agentes de Inteligencia Artificial conectarse nativamente para inspeccionar y controlar los entrenamientos de YOLO.
Instalación
Instala el paquete y sus dependencias (FastMCP) usando pip:
pip install -e .Related MCP server: Kubeflow MCP Server
Uso con un Cliente MCP (ej. Claude Desktop o Antigravity)
Añade este servidor a tu archivo de configuración MCP (por ejemplo, en mcp_servers.json):
{
"mcpServers": {
"neuralforge-mcp": {
"command": "wyolo-mcp",
"env": {
"NEURALFORGE_API_URL": "http://192.168.10.252:23442"
}
}
}
}Herramientas Expuestas
get_cluster_status: Comprueba si la API y Celery están online.get_study_details(study_id): Recupera métricas, progresos y el worker activo de un estudio.cancel_study(study_id): Fuerza la cancelación de un estudio y mata los contenedores efímeros.launch_training(config): Envía un nuevo estudio YOLO a la cola.check_dataset_path(path): Verifica localmente si un volumen o ruta de dataset existe antes de lanzar entrenamientos para evitar fallos.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/wisrovi/wyoloservice2_mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server