메뉴
Manage Vast.ai GPU instances — show status, start, stop, destroy, SSH, execute commands, view logs, copy files, take snapshots. Use when the user wants to check on, connect to, transfer files, or control their GPU instances.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | manage-instances |
| description | Manage Vast.ai GPU instances — show status, start, stop, destroy, SSH, execute commands, view logs, copy files, take snapshots. Use when the user wants to check on, connect to, transfer files, or control their GPU instances. |
| argument-hint | [action or instance-id] |
| allowed-tools | Bash |
Help the user manage their GPU instances on Vast.ai.
$ARGUMENTS
vastai show instances # Table view
vastai show instances --raw # JSON
vastai show instance <ID> --raw # Single instance JSON
vastai start instance <ID>
vastai stop instance <ID>
vastai reboot instance <ID> # Restart, keeps GPU priority
vastai recycle instance <ID> # Destroy + recreate fresh
vastai destroy instance <ID> # PERMANENT — confirm first
Batch: start instances, stop instances, destroy instances take multiple IDs.
Always confirm destroy with the user — it is irreversible.
vastai label instance <ID> '<LABEL>'
vastai update instance <ID> --label '<NEW>'
vastai bid instance <ID> --price <$/hr> # Change bid price
vastai prepay instance <ID> <AMOUNT> # Prepay credits
vastai ssh-url <ID>
# Returns: ssh://root@<HOST>:<PORT>
# Connect with: ssh -p <PORT> root@<HOST>
vastai scp-url <ID>
# Upload: scp -P <PORT> file.py root@<HOST>:/root/
# Download: scp -P <PORT> root@<HOST>:/root/results.tar.gz ./
vastai copy local:./data C.<ID>:/workspace/data # Upload
vastai copy C.<ID>:/workspace/results local:./results # Download
vastai copy C.<ID1>:/data C.<ID2>:/data # Instance-to-instance
vastai cancel copy C.<ID>:/path # Cancel transfer
vastai execute <ID> 'ls -l /workspace' # ls, rm, du only
vastai logs <ID> # Last 1000 lines
vastai logs <ID> --tail 50 # Last N lines
vastai logs <ID> --filter 'ERROR' # Filter lines
vastai logs <ID> --daemon-logs # System logs
vastai snapshot instance <ID> --repo <DOCKERHUB_REPO> [--tag <TAG>]
# Schedule reboots
vastai reboot instance <ID> --schedule DAILY --hour 3
# Schedule bid changes
vastai bid instance <ID> --price 0.5 --schedule WEEKLY --day 1
# View/delete scheduled jobs
vastai show scheduled-jobs
vastai delete scheduled-job <ID>
| Status | Meaning |
|---|---|
created | Just created, initializing |
scheduling | Waiting for resources |
running | Active and accessible |
exited | Container exited (check logs) |
stopped | Paused by user |
offline | Machine went offline |
When showing instances, note:
Manage Vast.ai autoscaling endpoints and worker groups for production deployments. Use when setting up auto-scaling GPU inference, managing worker pools, or deploying services.
Launch a GPU instance on Vast.ai. Use when the user wants to create, start, or spin up a new GPU machine for training, inference, or development.
Manage Vast.ai persistent storage volumes — search, create, delete, clone, and attach volumes to instances. Use for persistent data across instance lifecycles.
Run a job on a Vast.ai GPU instance end-to-end: find a GPU, launch it, execute the job, monitor it, and destroy the instance when done. Use when the user wants to run a training job, inference task, script, or any workload on a remote GPU.
Search for available GPU machines on Vast.ai. Use when looking for GPUs to rent, comparing GPU pricing, or finding machines that match specific requirements.
Set up Vast.ai CLI — configure API key, verify installation, set up SSH keys, and check account. Use when the user needs to get started with Vast.ai or is having authentication issues.