runpod-job

Orchestrate Runpod GPU tasks via transcript-backed SSH sessions with a pod_id.

11|1|Updated Feb 23, 2026

Search Tags:#automation #cloud #gpu #ssh #workflows #manual-workflow #runpod

Ready to run

npx skills add https://github.com/jimezsa/opencolab --skill runpod-job

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: runpod-job
Source: https://github.com/jimezsa/opencolab/tree/main/projects/SKILLS/runpod-job
Command: npx skills add https://github.com/jimezsa/opencolab --skill runpod-job

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps teams manage Runpod GPU tasks with a manual, human-in-the-loop workflow instead of automatic provisioning, by guiding the user to supply a pod_id, SSH access details, and a saved profile, then executing the GPU work through transcript-backed SSH sessions.

Core Features & Use Cases

Manual Runpod Pod workflow: the agent coordinates a user-created Pod, SSH access, and a stored profile to run GPU tasks.
Safe, bounded operations: all commands are executed via bounded SSH sessions rather than open-ended shells.
OpenColab-managed path on demand: users can opt into an OpenColab-managed lifecycle for run_id tracking when needed.

Quick Start

Provide the pod_id and SSH details to start a manual Runpod Pod workflow.

Frequently Asked Questions about runpod-job

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I manage Runpod GPU jobs manually via SSH?▼

To manage Runpod GPU jobs manually via SSH, you provide a pre-created pod_id and SSH access details to establish a human-in-the-loop workflow. The agent then executes your GPU tasks through bounded, transcript-backed SSH sessions for safe operations.

What is a manual Runpod Pod workflow and when should I use it?▼

A manual Runpod Pod workflow is a process where you control GPU resources by supplying a pod_id and SSH details rather than relying on automatic provisioning. Use it when you prefer human-in-the-loop control over pre-created Pods for executing GPU tasks.

Do I need to pre-create my Pod before starting a manual Runpod SSH workflow?▼

Yes, you must pre-create your Pod and provide the specific pod_id and SSH access details. This Skill enforces a manual workflow by requiring these pre-existing resources instead of automatically provisioning new GPU instances.

Can I use OpenColab-managed lifecycle tracking with Runpod GPU tasks?▼

Yes, you can opt into an OpenColab-managed lifecycle for run_id tracking on explicit request. By default, the workflow uses transcript-backed SSH sessions, but the managed path is available when you need automated run tracking.

Why does my Runpod agent use bounded SSH sessions instead of open-ended shells?▼

Your Runpod agent uses bounded SSH sessions to ensure safe, controlled operations during manual GPU task execution. This prevents open-ended shell access, keeping the human-in-the-loop workflow secure and limiting commands to the defined session scope.