← All skills

gke-ai-troubleshooting-tpu-performance-degradation

97 / 100 · Low risk

Fine to try. Some of what it can do is expected: a devops skill usually runs commands, installs tools and uses cloud credentials.

By google/skills (GitHub) · devops · 24,178 stars · scanned at commit 67c4dbc

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/*`). Distinguishes hardware and network fabric throttling from workload resource bottlenecks (HBM capacity, host memory, or host CPU saturation). Use when TPU training throughput or d…

Works with any agent that supports SKILL.md skills.

What this skill can do

Install

Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.

if [ -e ~/.agents/skills/'gke-ai-troubleshooting-tpu-performance-degradation' ] || [ -L ~/.agents/skills/'gke-ai-troubleshooting-tpu-performance-degradation' ]; then echo 'A skill folder named gke-ai-troubleshooting-tpu-performance-degradation already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/google/skills' "$d" && git -C "$d" checkout --quiet '67c4dbc8ff6313ef117d938e53bb21315fc281ae' -- 'skills/cloud/gke-ai-troubleshooting-tpu-performance-degradation' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/cloud/gke-ai-troubleshooting-tpu-performance-degradation' ~/.agents/skills/'gke-ai-troubleshooting-tpu-performance-degradation' && rm -rf "$d"; fi

Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'gke-ai-troubleshooting-tpu-performance-degradation' ~/.claude/skills/'gke-ai-troubleshooting-tpu-performance-degradation' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.

Badge for your README

The badge for google/skills shows the score of its lowest-scoring skill.

[![SkillTrust](https://zalavijaysinh.github.io/skilltrust/badges/google/skills.svg)](https://zalavijaysinh.github.io/skilltrust/repos/google/skills/)

Score history