runpod/runpod-plugins-official
8 skills89.4k installs50 starsGitHub
Install
npx skills add https://github.com/runpod/runpod-plugins-officialSkills in this repo
1RunpodctlProvides RunPod infrastructure management via the runpodctl CLI for GPU pods, serverless endpoints, and cloud resources. It supports deploying and operating ML workloads on RunPod from agent-driven terminals. Use when managing RunPod pods, endpoints, or GPU infrastructure programmatically.16.7kinstalls2RunpodProvides guidance for using the official RunPod plugin to deploy and manage GPU-backed AI workloads. Covers RunPod serverless endpoints, pod lifecycle, and integrating inference into agent or application workflows. Use when setting up RunPod infrastructure, configuring endpoints, or troubleshooting RunPod deployments for ML and AI services.15.1kinstalls3Runpod UsageHelps monitor and interpret RunPod usage, billing, and endpoint performance metrics. Covers cost optimization, scaling decisions, and identifying underused GPU resources. Use when tracking RunPod spend, right-sizing endpoints, or troubleshooting usage anomalies after deployment.15kinstalls4Runpod McpRunpod MCP skill for managing cloud GPU infrastructure through structured tool calls. Covers pods, serverless endpoints, jobs, templates, network volumes, container-registry auth, GPU/CPU catalog browsing, and billing via the Runpod MCP server integration.15kinstalls5Runpod MigrateA migration skill that moves a codebase from Runpod's GraphQL API or REST v1 to REST v2. It scans which code uses which version, rewrites the call sites, and reports breaking changes before verifying the result. A solo builder reaches for it when their GPU jobs run on the old Runpod API and they want the newer catalog and lifecycle states.13.3kinstalls6Runpod TemplatesThis skill is the reference for Runpod's official prebuilt pod templates, such as ComfyUI and PyTorch images. It explains what each image contains, which ports and paths it uses, how to pin a version, and what is missing on first boot. A solo builder reaches for it when a task says to run a known workload on a pod and an official template already covers it.12.6kinstalls7RunpodctlThe runpodctl skill Runpod CLI to manage your GPU workloads. Documentation covers workflows, commands, and guardrails agents should follow when users invoke this capability. Key documented areas include Use Hub when the user wants a known deployable app or worker such as vLLM, ComfyUI, Whisper, or a Runpod-maintained repo.; Use templates when the user already has a template ID, wants reusable image/config defaults, or needs lower-level control than Hub.; Use direct pod creation with `--image` when the user has a specific Docker image and does not need a saved template.; Use serverless for request/response inference APIs and scalable workers; use pods for interactive work, notebooks, training, debugging, or long-lived sessions. Reference commands include curl -sSL https://cli.runpod.net | bash; brew install runpod/runpodctl/runpodctl. Use when developers or agents need structured guidance for runpodctl tasks with evidence grounded in the bundled SKILL.md rather than generic advice.1.7kinstalls8RunpodOfficial Runpod plugin providing agent-first skills for GPU workload management. Manage pods, serverless endpoints, job queues, templates, and volumes via Runpod CLI, MCP server, and flash (custom code deployment). Six bundled skills cover MCP tools, CLI operations, flash deployment, and companion CLIs. OAuth-gated hosted MCP server included.60installs