Deployment and Operations Skills

Skills for standing up Dynamo deployments and diagnosing them when they misbehave.
以 Markdown 格式查看

These skills act on a running or intended cluster: deploying a recipe, starting a router, validating the interconnect fabric, and troubleshooting unhealthy workloads.

Prompts that reach them: “Deploy the Qwen3 32B FP8 recipe on my cluster and smoke-test the endpoint,” or “My prefill pods are restarting after a rolling update. Find out why.”

SkillUse for
deploy-dynamo-recipeDeploy an assigned DynamoGraphDeployment and verify it with an OpenAI-compatible smoke test.
dynamo-interconnect-checkValidate NIXL, Unified Communication X (UCX), and NVIDIA Collective Communications Library (NCCL) readiness for disaggregated serving over remote direct memory access (RDMA) or NVLink.
dynamo-router-starterStart or patch round-robin, KV-aware, least-loaded, or device-aware routing and run endpoint checks.
troubleshoot-dynamoDiagnose failed or unhealthy deployments, pods, model caches, workers, endpoints, or benchmark jobs.