$ loading_
Safely manage SLURM cluster jobs, files, logs, and remote commands.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "secure-cluster-mcp" yet — see the docs or source repo.
Upload local train.py and requirements.txt to the cluster project directory, create a SLURM job using 1 GPU named bert-train, and keep returning queue status plus a summary of the latest logs.
Transfers files, creates and submits the job, then returns the job ID, queue status, and key log details.
Inspect failed SLURM job 48291 from yesterday, read stdout, stderr, and scheduler details, determine the failure cause, and suggest next-step fixes; if safe, resubmit the corrected job.
Provides root-cause analysis, key log excerpts, fix recommendations, and optionally the resubmission result.
Safely run read-only checks on the login node: inspect disk quota, current queue usage, my running jobs, and a summary of error logs from the last 24 hours, then compile a concise report.
Returns a structured health report with resource usage, warning signals, and recommended actions.
Connect to Slurm clusters to inspect jobs, queues, and HPC scheduling tasks.
Secure MCP server for authenticated record queries and integration management.
Safely lets AI agents use threat analysis and security operations tools.
Let AI handle secure remote SSH commands, transfers, and port forwarding.
Secure LLM and MCP tool interactions with zero-trust controls and policy enforcement.
Let AI securely search and perform Slack API actions with OAuth guardrails.