umich-foreseer
Official@umich-foreseer
Facilitates high-performance computing cluster management, experiment submission, and cross-cluster connectivity for academic research environments using Slurm and SSH protocols.
Agent Skills by umich-foreseer
Showing 7 vetted skills indexed across 1 GitHub repositories.
onboard
Detect clusters and configure Claude Code and Codex for new lab members.
slurm-debug
Diagnose Slurm job failures by analyzing sacct, scontrol, and logs.
harvest
Detect completed SLURM experiments and collect results into docs.
slurm-job
Generate Slurm sbatch scripts with partition, account, GPU, and validation directives.
migrate
Discover repository compute assumptions and present a migration plan for review.
submit-experiment
Automate SLURM experiment submission with structured naming and documentation artifacts.
connect
Automate cross-cluster SSH tunnel setup between Great Lakes and Lighthouse.
Frequently Asked Questions About umich-foreseer
FAQPage SchemaWhat specific tasks can researchers perform using umich-foreseer?▼
Researchers can manage the full lifecycle of HPC experiments, including generating validated sbatch directives, submitting jobs with structured documentation, monitoring cluster performance via log analysis, and migrating repository compute assumptions between research environments.
Which technical personas benefit from these capabilities?▼
These capabilities are designed for computational researchers, lab members, and HPC system administrators who require standardized methods for managing Slurm-based workloads and maintaining secure connectivity between institutional research clusters.
What are the prerequisites for deploying these cluster management functions?▼
Deployment requires active access to the Great Lakes or Lighthouse research clusters and existing credentials for SSH-based network authentication. Users must also have the Slurm environment configured to interpret standard sbatch directives and sacct diagnostic logs.