Loading...
Loading...
Found 6 Skills
Monitors and troubleshoots GKE TPU workloads, nodes, and node pools using GKE system metrics and PromQL. Use when monitoring TensorCore duty cycle, TPU memory, node readiness, multi-host TPU node pool availability, host maintenance or preemption interruptions, and calculating MTTR or MTBI metrics for GKE TPUs. Don't use for general non-TPU GKE workload monitoring or non-metric TPU debugging.
Monitors and troubleshoots GKE TPU workloads, nodes, and node pools using GKE system metrics and PromQL. Use when monitoring TensorCore duty cycle, TPU memory, node readiness, multi-host TPU node pool availability, host maintenance or preemption interruptions, and calculating MTTR or MTBI metrics for GKE TPUs. Don't use for general non-TPU GKE workload monitoring or non-metric TPU debugging.
Personalized audit that teaches advanced Superset features the user isn't using yet — automations, parallel agents, tasks, multi-host, terminals, custom commands, MCP. Use when the user wants to get more out of Superset, learn advanced Superset features, or 10x their Superset workflow.
Decide whether to take a technical audience into an organization the organizer does not control, and on what terms - when a host's own premises, operation and staff are the programme rather than a room rented and filled by the organizer. Sets the host-dependency posture (host speaker hosted in-house, one host site host-led or to a negotiated agenda, a multi-host circuit) and the access posture, meaning what the host lets attendees see and repeat afterwards. Use whenever asked to design a learning expedition, company tour, field trip, site-visit programme or executive immersion. Do NOT use for catering - use samber/dev-event-organizer-skills@event-hospitality - or multi-venue logistics - use samber/dev-event-organizer-skills@event-production. States no legal threshold or instrument.
Diagnose GKE TPU VM instances stuck in Compute Engine repair (REPAIRING state) after host errors, maintenance events, or faulty host reports in single-host, multi-host, dynamic slicing, and TPU All Capacity mode node pools. Use when TPU instances in a slice node pool remain in REPAIRING, host error events appear in audit logs, or TPU slices are unschedulable due to a VM in repair. Don't use for nodes reporting Ready: Unknown when no VM is in REPAIRING (use gke-ai-troubleshooting-node-unresponsive-timeout), other NotReady nodes when no VM is in REPAIRING (use gke-node-notready), Pending Pods when all nodes are Ready (use gke-workload-troubleshooting), JobSet preemption events (use gke-ai-troubleshooting-jobset-interruption), or Slice custom resource lifecycle failures (use gke-ai-troubleshooting-tpu-dynamic-slices-monitoring).
[Bridging] Symlink a skill or an entire skills/ directory to every Agent host installed on this machine (public entry ~/.agents/skills, dedicated entries for Claude Code, WorkBuddy, Trae, etc.; generates a thin bridge for Grok). Also handles checking bridge status and removing bridges. Use when the user says 'bridge this skill', 'hook the skill into Codex/Doubao/WorkBuddy', 'make it available to all Agents', 'is it bridged yet?', 'cancel the bridge', 'unbridge'. | Crafted by XY