Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/gke-labs/kube-agentsnpx agentmods add skills/gke-labs/kube-agents/gke-stall-detectionWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/gke-labs/kube-agents/gke-stall-detection)<a href="https://agentmods.dev/skills/gke-labs/kube-agents/gke-stall-detection"><img src="https://agentmods.dev/badge/skills/gke-labs/kube-agents/gke-stall-detection/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/gke-labs/kube-agents/gke-stall-detection"><img src="https://agentmods.dev/badge/skills/gke-labs/kube-agents/gke-stall-detection.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00048 | $0.02124 |
| Opus 5.5 | $0.00019 | $0.00850 |
| Sonnet 5.5 | $0.00010 | $0.00425 |
| Haiku 4.5 | $0.00005 | $0.00212 |
Grade A, and why
gke-stall-detection scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 16d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GKE Stall Detection
Use this skill when a controller has gone quiet rather than red: a Gateway that never programs, a rollout that never finishes, a custom resource whose status stopped following its spec, an HTTPRoute nobody accepted. The symptom-driven skills (gke-workload-troubleshooting, gke-reliability) start from a failing pod or a bad event; this one starts from the absence of progress, which those states never show. It is read-only: it lists objects, reads their status and events, resolves referenced names, and proposes nothing beyond naming what is missing.
When to run it
- A user reports something "stuck", "still pending", "never came up", or "not reconciling" and nothing is CrashLooping.
- A Gateway, HTTPRoute, Certificate, Config Connector or Argo-managed object has been in the same state for longer than its controller should need.
- Before declaring a namespace healthy in a reliability review: a silent stall is what that review otherwise misses.
Heuristics and thresholds
/opt/data/scripts/stall_report.py applies four heuristics to every object it reads. Each is gated by an age threshold, so a controller that is merely slow is not reported.
| Heuristic | Fires when | Age measured from |
|---|---|---|
generation-lag |
metadata.generation is ahead of status.observedGeneration (objects without observedGeneration are skipped) |
newest managedFields entry that wrote f:spec |
stale-condition |
A Progressing, Accepted, Programmed, ResolvedRefs, Ready or Available condition is not True, including conditions nested under status.listeners[] or status.parents[] |
the condition's lastTransitionTime |
repeating-warnings |
A Warning event on the object has a count of at least 3, spans at least the threshold, and was last seen within the threshold | the event's first to last occurrence |
dangling-reference |
The spec names an object that does not exist: certificateRefs, configMapRef, configMapKeyRef, secretRef, secretKeyRef, parentRefs, backendRefs, and volume configMap/secret/persistentVolumeClaim sources; optional: true references are skipped |
newest managedFields entry that wrote f:spec |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 16d ago Changed · +3 lines b72af6549b32
- 1mo ago First seen · 83 lines · 48 tokens per session scan A 428de6a3b1a5
gke-stall-detection is a skill published in the GitHub repository gke-labs/kube-agents (109 stars, last pushed today), licensed Apache-2.0. It adds 48 tokens to every session and 2,124 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-15.
Other skills, from other repositories
gke-workload-troubleshooting
Diagnoses GKE workload failures (CrashLoopBackOff, OOMKilled, ImagePullBackOff, Pending, etc.) via logs and events. Use when pods fail to start or crash repeatedly. Don't use for GKE cluster infrastructure provisioning, node pool creation, or non-Kubernetes Google Cloud services.
gke-workload-scaling-troubleshooting
Diagnoses GKE HorizontalPodAutoscaler (HPA) failures — metrics showing as , FailedGetResourceMetric / FailedGetScale / FailedComputeMetricsReplicas events, missing Pod resource requests, custom/external metrics-pipeline breakage (FailedGetExternalMetric / FailedGetCustomMetric, unavailable metrics adapter…
gke-ai-troubleshooting-tpu-mxla-hang
Diagnose GKE Cloud TPU multi-slice training hangs (Megascale HANGDETECTED logs and hang monitored events) using the ML Diagnostics Megascale XLA (MXLA) Hang Analyzer (gcloud alpha mldiagnostics monitored-events) and 1-minute Cloud Monitoring multi-slice latency metrics (kubernetes.io/container/multislice/).…
gke-ai-troubleshooting-tpu-performance-degradation
Diagnose GKE Cloud TPU training throughput drops and step-time regressions (15%+ TPU duty-cycle drop) using ML Diagnostics Workload Monitoring (gcloud alpha mldiagnostics monitored-events / hypercomputecluster.googleapis.com/v1alpha) and 1-minute Cloud Monitoring system metrics (kubernetes.io/node/accelerator/).…
gke-ai-troubleshooting-jobset-interruption
Diagnoses GKE JobSet interruptions, restarts, and preemptions for AI/ML training workloads autonomously. Use when troubleshooting JobSet restart loops, spot VM preemptions, node readiness failures, host VM issues, or coordinator worker crashes. Don't use for general GKE cluster creation, basic workload deployment, or…
gke-node-notready
Diagnoses GKE nodes reporting NotReady or Unknown status by inspecting node conditions, events, kubelet/containerd logs, and node metrics, then proposing safe remediations. Use when nodes show NotReady, when the kubelet stops posting node status, or when workloads are evicted or stuck Pending due to node health. Don't…