hpc-monitor
Skill36 starsChanged 6 months agoMITView on GitHub
Use when the user wants a one-shot read-only status view of their compute — local machine + SLURM GPU cluster + a scheduler-less CPU server + Matlantis (papermill MD jobs) — as tqdm-style bars. Pure bash + ssh, no LLM/API. Triggers - hpcmon, HPC 모니터, GPU 현황, 큐 상황, squeue, 서버 현황, 잡 현황, matlantis 잡, MD 잡 확인, "얼마나 busy".
In this repository29 other files it gives its agents.
- skills/hpc-monitor/AGENTS.mdAGENTS.md
- aseSkill
- blender-atom-renderSkill
- catagentSkill
- ccel_slide_skillSkill
- computational-section-writingSkill
- design2htmlSkill
- docx-scientific-formattingSkill
- figure_constructionSkill
- html-minimalSkill
- humanizer_engSkill
- humanize-writingSkill
- ko-writingSkill
- rxn-diagram-insetSkill
- matplotlib-scientificSkill
- meetingnote-paperworkSkill
- openrouter-costSkill
- ovito-atom-renderSkill
- paper-bibSkill
- paperSkill
- pptx-too-heavySkill
- report_less_slopSkill
- reportSkill
- slide-auditSkill
- smart-compactSkill
- soagentSkill
- transcript2htmlSkill
- youtube2mp4Skill
- youtubeSkill
The file
Files
hpc-monitor
로컬 머신 + 원격 3종(SLURM GPU 클러스터 · 스케줄러 없는 CPU 서버 · Matlantis 인스턴스)의 사용률·잡 현황을 한 번에 뽑는 읽기전용 bash 모니터. LLM/네트워크 API 없이 순수 bash + ssh 로 동작하며, 원격에서 실행하는 명령은 전부 관측용(squeue/sinfo/ps/free/uptime/grep/awk)이라 서버 상태를 바꾸지 않는다.
구성
monitor_all.sh— 오케스트레이터. 아래 3개를 병렬 SSH로 던지고 취합.hpcmonalias 대상.gpu_status.sh— SLURM GPU 점유/실행/대기 현황 (로그인 노드에서 실행).mace_status.sh— 스케줄러 없는 CPU 서버 load/RAM/계산 프로세스.mtl_status.sh— Matlantis papermill 백그라운드 잡 현황 + "내 잡" 진행률.
Usage
# 1) monitor_all.sh 상단 config (SLURM_USER, MTL_PREFIX, HPC_HOST, CPU_HOST, MTL_HOST) 를 본인 환경으로 편집
# 2) alias 등록 후:
hpcmon개별 실행:
ssh "$HPC_HOST" "bash -s -- $SLURM_USER" < gpu_status.sh
ssh "$CPU_HOST" 'bash -s' < mace_status.sh
ssh "$MTL_HOST" "bash -s -- $MTL_PREFIX" < mtl_status.sh셋업 상세는 README.md, 에이전트용 안내는 AGENTS.md 참고.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

