google/skills is a collection of Agent Skills that guide AI agents in using Google products and technologies, including Google Cloud services. Developers and agents use these reusable workflows for tasks such as cloud authentication, data analysis, application deployment, and AI platform management. The catalogue entries are skills, plugins, and an MCP integration from this collection.
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 google/skills --skill gke-ai-troubleshooting-tpu-reservation-stockoutgit clone --depth 1 https://github.com/google/skillsWrote 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/google/skills/gke-ai-troubleshooting-tpu-reservation-stockout)<a href="https://agentmods.dev/skills/google/skills/gke-ai-troubleshooting-tpu-reservation-stockout"><img src="https://agentmods.dev/badge/skills/google/skills/gke-ai-troubleshooting-tpu-reservation-stockout/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/google/skills/gke-ai-troubleshooting-tpu-reservation-stockout"><img src="https://agentmods.dev/badge/skills/google/skills/gke-ai-troubleshooting-tpu-reservation-stockout.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 | $0.00138 | $0.04128 |
| Opus 5.5 | $0.00055 | $0.01651 |
| Sonnet 5.5 | $0.00028 | $0.00826 |
| Haiku 4.5 | $0.00014 | $0.00413 |
Grade A, and why
gke-ai-troubleshooting-tpu-reservation-stockout 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 5d 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 — 291 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Troubleshoot GKE TPU reservation stockouts caused by leaked node pool capacity
Use this skill to diagnose and resolve GKE TPU node pool creation or scale-up
failures that fail with GCE_STOCKOUT, ZONE_RESOURCE_POOL_EXHAUSTED, or
ZONE_RESOURCE_POOL_EXHAUSTED_WITH_DETAILS (or queue due to lack of capacity)
even though they target a specific Compute Engine TPU reservation.
If node auto-repair is disabled on an existing TPU node pool, GKE doesn't repair
its nodes when they become NotReady. While their VMs are still running, they
keep consuming slots in the reservation (inUseCount is the number of running
instances that consume the reservation). Per
Ensure reservation availability,
the reservation must have sufficient available TPU chips at the time of node
pool creation, and per
Multi-host TPU slice node pools,
if GKE is unable to deploy one node in a slice, no nodes in the TPU slice node
pool are deployed. When the free slots in the reservation
(specificReservation.count - specificReservation.inUseCount; see
Consumable instances in a reservation)
are fewer than the number of nodes that the new TPU node pool needs, creation
fails with GCE_STOCKOUT or ZONE_RESOURCE_POOL_EXHAUSTED (or queues due to
lack of capacity).
Attribution
MUST READ: Before executing any gcloud, REST API, or MCP command, you
MUST read Usage Attribution. For commands executed
directly in the terminal (never in generated scripts, Terraform, or user code),
prefix gcloud with
CLOUDSDK_METRICS_REQUEST_ATTRIBUTION="agent-skills/1.0 (skill:gke-ai-troubleshooting-tpu-reservation-stockout)"
and append agent-skills/1.0 (skill:gke-ai-troubleshooting-tpu-reservation-stockout) to User-Agent headers.
Prerequisites
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 5d ago Changed · +9 lines cc30a83b8317
- 9d ago First seen · 282 lines · 138 tokens per session scan A 063dd42f12fb
gke-ai-troubleshooting-tpu-reservation-stockout is a skill published in the GitHub repository google/skills (21,557 stars, last pushed yesterday), licensed Apache-2.0. It adds 138 tokens to every session and 4,128 once invoked, about $0.0006 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 2026-10-16.
Other skills, from other repositories
scheduled-jobs
Covers cron and scheduled work done right — idempotency, preventing overlapping runs, monitoring for missed and failed runs, alerting on silence, and getting time zones and DST transitions correct. Use this whenever the user is writing a cron job or Kubernetes CronJob, asks why a scheduled job ran twice or didn't run…
configuration-management
Covers Ansible, Chef, and Puppet for managing mutable systems declaratively — idempotent tasks, convergence toward desired state instead of one-off scripts, inventory organization, and roles. Use this whenever the user is writing a playbook, cookbook, or manifest, debugging why a run isn't converging, organizing hosts…
immutable-infrastructure
Covers replacing servers wholesale instead of patching them in place — baking golden images, treating instances as disposable cattle rather than nursed pets, rebuilding to make any change, and the rollback simplicity that buys. Use this whenever the user is deciding whether to SSH into a running instance to fix it…
infrastructure-as-code
Treats infrastructure changes as version-controlled, reviewable configuration instead of manual clicks or SSH sessions — Terraform state, plan review, environment parity, and guardrails against irreversible changes. Use this whenever the user is writing or reviewing Terraform, CloudFormation, or Pulumi code, applying…
autoscaling
Covers scaling Kubernetes workloads and nodes to demand — HPA on the right metric, VPA, cluster autoscaler, custom/external metrics, avoiding thrash with stabilization windows, and requests as the foundation underneath it all. Use this whenever the user is configuring an HPA, deciding between HPA and VPA, debugging…
kubernetes-networking
Explains how traffic reaches and moves between pods — Services (ClusterIP/NodePort/LoadBalancer), Ingress and controllers, cluster DNS, NetworkPolicy default-deny, and debugging selector mismatches or empty endpoints. Use this whenever the user asks why a Service has no endpoints, why a request 502s inside the…