ksterx/srunx

What it does

ksterx/srunx is an MCP (Model Context Protocol) server that bridges natural language interfaces with SLURM (Simple Linux Utility for Resource Management), the industry-standard workload manager for high-performance computing clusters. It enables users to submit jobs, execute YAML workflows, monitor GPU resources, manage SSH profiles, and synchronize files to remote HPC clusters using conversational commands. The tool provides 14 specialized tools spanning both local SLURM and SSH-enabled remote cluster operations.

Key features

• Job submission and management through natural language
• YAML workflow execution and orchestration
• Real-time GPU resource monitoring and status tracking
• SSH profile management for multi-cluster access
• File synchronization to remote HPC systems
• Companion CLI tool for terminal-based operations
• FastAPI-based web UI for browser access
• Support for both local and remote cluster operations

How to set up

Installation requires Python and Git. Clone the repository from GitHub, install dependencies via pip, configure your SLURM cluster access and SSH profiles, then launch the MCP server or web UI component. The package includes integrated CLI and FastAPI web interface for immediate use after setup.