206 件 ・ 関連度順
概要と使いどころ
Triage OpenSEO production errors in Cloudflare Workers Observability — verified query recipes, counting gotchas, and a known-noise filter list applied automatically. Use when asked to review Cloudflare logs/observability, count OOMs or worker errors, compare error rates between periods, or investigate a prod error spike.
日本語の概要は準備中です。原文の説明を表示しています。
Mounts Cloud Storage buckets on Google Cloud as a POSIX file system using Cloud Storage FUSE (and the gcsfuse CLI tool). Use when interacting with the gcsfuse CLI or Cloud Storage FUSE mounts: decide whether FUSE, native gs:// reads, or Filestore/Managed Lustre fits a workload, deploy tuned mounts on GKE, Compute Engine, or Cloud Run, enable and size file, stat, and list caches, tune mount flags (--implicit-dirs) or config-file settings, apply workload profiles, keep ML checkpointing safe (rename atomicity, hierarchical namespace/HNS, close-time finalization, concurrent writers), or diagnose slow training, low throughput, or bill spikes with gcsfuse metrics. Covers mount semantics, gcsfuse CLI and config files, GKE gcsfuse CSI driver (Workload Identity principal:// bindings, profile StorageClasses, sidecar sizing), and Cloud Run volume mounts. Don't use for bucket administration or data management without a mount (google-cloud-storage-basics) or fully POSIX-compliant shared file systems (Filestore, Managed Lustre; use gke-storage).
日本語の概要は準備中です。原文の説明を表示しています。
Provides diagnostic workflows and step-by-step root-cause analysis procedures for actively broken, failing, or slow BigQuery jobs, execution graph and query plan stage bottlenecks, system performance issues, or unexpectedly expensive workloads. Use when interpreting symptoms, isolating bottlenecks, diagnosing cost spikes (on-demand query spend, capacity slot autoscaling, storage growth), execution graph stages or substep variables, identifying root causes, and determining remediation steps. Don't use for writing or optimizing SQL, proactive capacity planning, or storage layout design (use bigquery-optimization), or when the user already knows which telemetry they want and just needs the query (use bigquery-observability).
日本語の概要は準備中です。原文の説明を表示しています。
Periodic sweep that finds incidents marked resolved in {{incident_channel}}, correlates the channel timeline against GitHub deploys/merges and Datadog log spikes over the incident window, and drafts a structured postmortem as a PR under {{postmortem_path}} in {{target_repo}} — root cause and action items proposed, never finalized.
日本語の概要は準備中です。原文の説明を表示しています。
On-call alert triage runbook for {{sentry_project}}. Pulls the Sentry stack trace, correlates Datadog logs around the spike, checks {{target_repo}} for the deploy that shipped just before, classifies severity against known-noisy patterns, posts a first-pass diagnosis to {{incident_channel}}, and pages a human only when severity is at or above {{page_severity_threshold}} or the alert can't be resolved.
日本語の概要は準備中です。原文の説明を表示しています。
Field-tested methodology and concrete recipes for training and operating large-scale LLM/VLM/multi-modal models end to end - choosing and benchmarking accelerators, storage and network; SLURM/Kubernetes orchestration; maximizing training throughput and fitting models in memory; diagnosing and surviving training instabilities, NaN/Inf, and hardware/job failures; checkpointing and fault tolerance; inference performance and memory; debugging multi-node/ multi-GPU hangs; and writing/running tests. Use when the user is training or fine-tuning large models, hits low TFLOPS/MFU, OOM, slow dataloading, a loss spike/divergence, a NCCL/InfiniBand or multi-node hang, node/GPU failures, checkpoint or preemption problems, storage/network bottlenecks, or needs to pick GPUs/cloud/file-systems or size inference latency/throughput. Distilled from "Machine Learning Engineering", the latest version of which can be found at https://github.com/stas00/ml-engineering The latest SKILL.md version can be found at https://github.com/stas00/ml-engineering/blob/master/skills/ml-engineering/SKILL.md
日本語の概要は準備中です。原文の説明を表示しています。
Investigate an OpenShell problem and create a structured issue with technical findings for human disposition.
日本語の概要は準備中です。原文の説明を表示しています。
Battle-tested PyTorch training recipes for all domains — LLMs, vision, diffusion, medical imaging, protein/drug discovery, spatial omics, genomics. Covers training loops, optimizer selection (AdamW, Muon), LR scheduling, mixed precision, debugging, and systematic experimentation. Use when training or fine-tuning neural networks, debugging loss spikes or OOM, choosing architectures, or optimizing GPU throughput.
日本語の概要は準備中です。原文の説明を表示しています。
exploration capture | explore sketch spike spec capture
日本語の概要は準備中です。原文の説明を表示しています。
Run or plan ChIP-seq, CUT&RUN, or CUT&Tag QC, control handling, spike-in, peak calling, broad-vs-narrow target selection, replicate, bigWig, and differential binding workflows.
日本語の概要は準備中です。原文の説明を表示しています。
Build the effects of a water and crystal skill as one visual language in Three.js. Water that reads as clear water, never a teal fill or a textured ball. A ray-marched water orb whose surface swells, rings and deforms in place: it refracts the frame behind it with an inverted horizon, has a silver rim and sharp highlights, and shows a faint caustic web on its far wall. Streams wound up from tide pools as uneven beaded ropes with silver edges, drips and shed drops. A lash that unspools from the orb, whips and pours into a pool under a torn, lopsided splash crown of aerated white water with spikes breaking into drops. Analytic ripple rings and rain on the pools. Crystals traced against their own half-space planes (Snell, Fresnel, total internal reflection, exit dispersion, Beer–Lambert, fracture discs, a white-hot root glow and a resonance band that climbs the stone), growing out of the water in a draining veil, then shattering into Voronoi shards that fall, tumble and rest with real contact on a heightfield. Caustics on the pool floor, and a moonlit tidal rock flat: rock banks traced from target images onto a world grid, flat mirror pools, a moon-glitter column of micro-facet glints, and a moon disc with maria. Use for water or ice-crystal skills, water orbs or bubbles, water whips, jets or tentacles, splash impacts, tide pools, rain, crystal growth or eruption, crystal resonance, shattering glass or quartz, caustics, wet-stone night scenes, moon reflections on water, "water and crystal", "hydro", "clear water VFX", "crystal shatter" or "skill effects only". Includes a reusable module and a five-beat demo on a scanned CC0 low-tide floor with failure-mode switches.
日本語の概要は準備中です。原文の説明を表示しています。
Defamiliarization audit for empirical output. Systematically interrogates every feature of a figure, table, or set of results — not just the main finding. Named for Jason Fletcher, who asked about the spike at t=1 when everyone else was looking at t=2. Use when you have output and are about to interpret or report it.
日本語の概要は準備中です。原文の説明を表示しています。
When your CPA spikes, Claude breaks down exactly what caused it. It looks across your campaign data and isolates the contributing factors — audience fatigue, bid landscape shifts, creative decay, landing page conversion drops, budget distribution changes, or new competitor activity. Platform: Google and Meta.
日本語の概要は準備中です。原文の説明を表示しています。
Catches unusual performance changes across your accounts — CPC spikes, CVR drops, spend surges, impression collapses, CTR shifts — and flags them with context about what likely changed. The goal is to catch problems in hours instead of discovering them days later during a routine check. Platform: Google and Meta.
日本語の概要は準備中です。原文の説明を表示しています。
Taking a multi-PR feature from "shape unknown" to merged without five review rounds per PR: spike until interfaces stop moving, write the contract, cut PRs along contract seams, one review per PR. Use when starting a feature that touches more than one lifecycle object or public interface, when a PR review keeps changing interfaces, or when splitting a large branch.
日本語の概要は準備中です。原文の説明を表示しています。
Investigate outliers, rare events, spikes, and suspicious records in datasets. Use as an explicit anomaly-analysis helper when you want concrete anomaly-detection workflow guidance, not generic data validation or end-to-end ML ownership.
日本語の概要は準備中です。原文の説明を表示しています。
[omh] Software slowness, memory leaks, or cost spikes: find where a system is actually slow, leaking, or expensive across runtime, memory, token cost, storage, rendering, inference, CI, and query domains, then fix one measured hot path at a time behind a regression budget. Use when the user says: ultraperf, ulw-perf, performance audit, performance bottleneck, find the bottleneck, profile the hot path, memory leak investigation, token cost hotspot.
日本語の概要は準備中です。原文の説明を表示しています。
Stock, capacity, and availability monitoring, acting on low-stock and slow-mover alerts, triaging order exceptions such as delays, return spikes, and complaints, and the operator's daily briefing; any start-of-day or what-needs-attention rundown is this flow, presented as a digest. Not needed for performance questions with no operational action attached.
日本語の概要は準備中です。原文の説明を表示しています。
Forecast Azure spend and price planned resources or workloads. WHEN: "forecast Azure spending", "project next month cost", "compare forecast to my planning target", "how much will Azure cost", "estimate VM cost", "compare storage tiers", "compare Azure regions", "retail price", "EA rate card", "negotiated pricesheet". DO NOT USE FOR: configured budget health or alerts, bill analysis, cost spikes, rightsizing, or commitments.
日本語の概要は準備中です。原文の説明を表示しています。
Optimize existing Azure resources and analyze Reservations or Savings Plans. WHEN: "optimize Azure costs", "reduce cloud spending", "rightsize resources", "find idle resources", "orphaned disk", "deleted VM still charged", "public IP still charging", "reservation utilization", "Savings Plan coverage", "commitment recommendation", "why is pay-as-you-go still charged". DO NOT USE FOR: cost spikes, forecasts, pricing estimates, budgets, or governance.
日本語の概要は準備中です。原文の説明を表示しています。
Analyze actual Azure spend, bill changes, and AKS or AI costs. WHEN: "Azure cost breakdown", "most expensive resources", "why did my bill increase", "cost spike", "unexpected charge", "Cosmos DB charges", "Foundry model costs", "AKS cluster cost", "namespace cost", "enable AKS cost analysis", "idle AKS capacity". DO NOT USE FOR: forecasts, pricing estimates, rightsizing, commitments, or budgets.
日本語の概要は準備中です。原文の説明を表示しています。
Use when the user asks to "build our social crisis protocol", "mentions are exploding — what do we do first", or "when do we pause the posting queue"; produces a 1-5 severity ladder with tunable Estimated trigger thresholds (mention-velocity multiples vs the 7-day listening baseline, sentiment flip, journalist/regulator contact, employee-conduct class), the first-mechanical-action rule — pause ALL scheduled posts AND paid amplification, with dated state markers dropped to the channels proposal protocol and reconciled post-incident — a pre-approved holding-statement library with committed update cadences, when-NOT-to-post rules, a spokesperson/approval matrix, and a post-crisis retro template; re-runs the social-quality-auditor pre-publish gate before un-pausing the queue. Not for email deliverability incidents (blocklist, spam-rate spikes) — use deliverability-qa; inside a launch window launch-day-conductor owns incident handling. 社媒危机预案/暂停队列/声明库/发言人矩阵
日本語の概要は準備中です。原文の説明を表示しています。
Use when the user asks to "keep the launch momentum going after launch week", "plan a changelog / release-notes cadence as GTM", or "is this update worth a relaunch"; produces a T+1→T+30 momentum plan — a launch-moment calendar (milestone / shipped-loop / badge moments only), announcement-tier routing (major = full-channel, medium = targeted, minor = changelog-only), a relaunch legitimacy call, spike-to-owned handoff briefs, and the next Tier-1 moment with launch-stacking spacing. Not for the 30-day content-reuse map or paid amplification execution — use content-amplifier; not for planning the next launch end to end — use launch-tier-planner. 抗第二周断崖/changelog-as-GTM/relaunch/下一发布时刻
日本語の概要は準備中です。原文の説明を表示しています。
Use when the user asks to "check pacing", "am I over/under-spending", "is this campaign on track to hit budget", or "why did spend spike/stall mid-flight"; returns a spend-vs-target-curve read, learning-phase status, an over/under-delivery call, and a reallocation trigger. Not for initial budget allocation — use budget-optimizer; not for choosing the bid strategy — use bid-strategy-planner; not for the RQS gate — use ad-account-auditor. 付费广告预算节奏监控/跑量过快过慢/在途配速
日本語の概要は準備中です。原文の説明を表示しています。