Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add nebius/nebius-physical-ai --skill debug-failed-rungit clone --depth 1 https://github.com/nebius/nebius-physical-aiWrote 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/nebius/nebius-physical-ai/debug-failed-run)<a href="https://agentmods.dev/skills/nebius/nebius-physical-ai/debug-failed-run"><img src="https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/debug-failed-run/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/nebius/nebius-physical-ai/debug-failed-run"><img src="https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/debug-failed-run.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00052 | $0.02879 |
| Opus 5.5 | $0.00021 | $0.01152 |
| Sonnet 5.5 | $0.00010 | $0.00576 |
| Haiku 4.5 | $0.00005 | $0.00288 |
Grade A, and why
debug-failed-run 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 19d 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 — 257 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Debug a failed workflow run
Start from the run id and work outward. Every step below is read-only until the final decision, so none of it costs GPU time or destroys the evidence you need.
0. Find the run id
npa workbench workflow list --project <alias> --limit 50 --json
Runs are durable S3 objects, so this works after the controller, the cluster, or
your shell is gone. If you only have an S3 prefix, every command below accepts an
s3:// run prefix wherever it accepts a run id.
1. Status first — it names the failure class
npa workbench workflow status <run-id> --project <alias> --json
This is the single most informative command, because for a PENDING job it reports the pod-level reason rather than just "pending". That distinction is the whole triage:
PENDING+ImagePullBackOff→ registry problem, go to step 4. The job will never fail on its own; Kubernetes retries pulls forever.PENDING+Unschedulable→ capacity or accelerator-name problem, step 5.FAILED→ the payload ran and exited non-zero, step 3.FAILED_STARTUP→ the controller itself failed identically several times (threshold--startup-failure-threshold, default 3). This is infrastructure, not your payload.
Use the run's selected --project consistently. Status, pending-manifest lookup,
live logs, and cancellation carry the resolved project and storage credentials into
their controller calls, including a cold restart of an owned local API. The
caller environment is unchanged. This does not permit adopting a controller
under a different principal or ignoring changed credential files; reconcile
those identity failures before resuming.
All currently recorded jobs can be SUCCEEDED between waves while the workflow
lifecycle remains RUNNING. This is a successful observation of incomplete
workflow evidence, not a failed query: inspect workflow_lifecycle for the
original durable timestamp, completion_recorded: false, and
driver_liveness: unknown. Fresh job polls do not prove driver liveness or create
progress heartbeats. Keep the original driver running and let --watch cross the
handoff; do not recommend resume solely from this state. Actual query failures
still exit nonzero, and terminal workflow/artifact proof remains required.
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.
- 19d ago Changed · +7 lines 3403ed298a7f
- 20d ago Changed · +9 lines 2c6ad87299f1
- 1mo ago Changed · +46 lines ec47aa1324e1
- 1mo ago First seen · 195 lines · 52 tokens per session scan A ad05d33c0789
debug-failed-run is a skill published in the GitHub repository nebius/nebius-physical-ai (50 stars, last pushed today), licensed Apache-2.0. It adds 52 tokens to every session and 2,879 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 2026-08-30.
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…