部署与运维 Skills - 部署与运维 Agent Skills推荐与安装 - 第20页 - AI导哈

gke-ai-troubleshooting-tpu-performance-degradationgoogle/skillsDiagnose 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 duty cycle drops without crashing pods, when `PERFORMANCE_DEGRADATION` monitored events fire, or when triaging slow multi-slice training steps. Don't use for complete multi-slice XLA execution stalls with `HANG_DETECTED` logs (use gke-ai-troubleshooting-tpu-mxla-hang) or pod eviction/interruption restarts (use gke-ai-troubleshooting-jobset-interruption).927 gke-ai-troubleshooting-tpu-mxla-hanggoogle/skillsDiagnose GKE Cloud TPU multi-slice training hangs (Megascale `HANG_DETECTED` 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/*`). Distinguishes XLA compiler/HLO launch divergence and host data-input stalls from TPU chip, SparseCore, ICI, or network fabric faults. Use when multi-slice TPU training jobs freeze without progressing steps, emit `HANG_DETECTED`, or stall in collective operations. Don't use for gradual step-time throughput drops without hangs (use gke-ai-troubleshooting-tpu-performance-degradation) or pod preemption/eviction restarts (use gke-ai-troubleshooting-jobset-interruption).927 senior-ml-engineeralirezarezvani/…e-skillsML engineering skill for productionizing models, building MLOps pipelines, and integrating LLMs. Covers model deployment, feature stores, drift monitoring, RAG systems, and cost optimization. Use when the user asks about deploying ML models to production, setting up MLOps infrastructure (MLflow, Kubeflow, Kubernetes, Docker), monitoring model performance or drift, building RAG pipelines, or integrating LLM APIs with retry logic and cost controls. Focused on production and operational concerns rather than model research or initial training.925 infisical-secret-syncsinfisical/ai-skillsGuide for configuring Infisical Secret Syncs to push secrets from Infisical to third-party services. Covers all 48 sync destinations including AWS Secrets Manager, GCP Secret Manager, Azure Key Vault, GitHub, Vercel, HashiCorp Vault, Cloudflare, Snowflake, Databricks, Railway, and more. Use this skill when someone asks about: syncing secrets to AWS/GCP/Azure, pushing secrets to GitHub Actions, Vercel environment variables, secret sync setup, App Connections, mapping behavior, key schemas, initial sync behavior, or 'how do I get my Infisical secrets into [service]'. Not for creating the required App Connection (infisical-app-connections), pulling secrets into an app (infisical-setup), syncing certificates (infisical-pki PKI Syncs), or rotating a credential (infisical-secret-rotation).920 cosmosdb-best-practicescline/skillsAzure Cosmos DB performance optimization and best practices guidelines for NoSQL, partitioning, queries, SDK usage, and vector search. Use when writing, reviewing, or refactoring code that interacts with Azure Cosmos DB, designing data models, optimizing queries, or implementing high-performance database operations.920 vercel-labs-emulatereason-machines/trending-skillsLocal drop-in API emulation for Vercel, GitHub, and Google services in CI and no-network sandboxes919 game-deployplayableintelligence/game-creatorDeploy browser games to here.now (default), GitHub Pages, or other hosting. Use when deploying a game, setting up hosting, or publishing a game build. Do NOT use for local development servers (use npm run dev).915 cloudflare-tunnelokou-ai/okou-skillsCloudflare Tunnel API for secure tunnels. Use when user mentions "Cloudflare tunnel", "argo tunnel", or secure connectivity.909 databricks-dabsdatabricks/data…t-skillsCreate, configure, validate, deploy, run, and manage Declarative Automation Bundles (DABs, formerly Databricks Asset Bundles). Use when working with Databricks resources via DABs including dashboards, jobs, pipelines, alerts, volumes, and apps.907 benchmark-modelmodular/skillsBenchmark a model served on MAX with the `max benchmark` command: measure throughput (tokens/sec), latency (TTFT, TPOT, inter-token latency), and GPU utilization by driving load against a running `max serve` endpoint. Use this whenever the user wants to benchmark, load-test, or measure the performance of a MAX model, get tokens-per-second / TTFT / TPOT numbers, run a concurrency or request-rate sweep, compare latency vs throughput, size a deployment, or produce benchmark JSON, even if they don't say "benchmark" by name. Also use when a `max benchmark` run fails to connect or reports zero/garbage numbers.903 pulumi-cdk-to-pulumipulumi/agent-skillsLoad this skill when a user wants to migrate, convert, port, translate, or move an AWS CDK application (including CDK stacks, constructs, or CloudFormation-synthesized templates) to Pulumi. Phrases such as "convert CDK to Pulumi", "migrate CDK app", "port CDK stacks", "replace CDK with Pulumi", "stop using CDK". Do NOT load for general CDK questions, CDK-only help, or CDK vs Pulumi comparisons where no migration is requested.903 iot-firmwareruvnet/rufloOrchestrate firmware rollouts with canary deployment and anomaly-gated advancement901 scikit-learn-best-practicesmindrally/skillsBest practices for scikit-learn machine learning, model development, evaluation, and deployment in Python897 sentry-setup-ai-monitoringgetsentry/sentr…t-skillsSetup Sentry AI Agent Monitoring in any project. Use when asked to monitor LLM calls, track AI agents, or instrument OpenAI/Anthropic/Vercel AI/LangChain/Google GenAI/Pydantic AI. Detects installed AI SDKs and configures appropriate integrations.896 cloudformation-to-pulumipulumi/agent-skillsConvert, migrate, or import AWS CloudFormation stacks or templates into Pulumi programs. Load this skill whenever a user wants to move from CloudFormation to Pulumi, convert a CFN template, import existing CloudFormation-managed resources into Pulumi, or asks about CloudFormation-to-Pulumi migration in any form. Also load when the user mentions cdk-importer in a migration context.894 nemoclaw-setupjezweb/claude-skillsInstall and configure NVIDIA NemoClaw (sandboxed OpenClaw agent platform) on Linux. Handles cloudflared tunnels, Docker cgroup fixes, OpenShell, sandbox creation, remote access via Cloudflare Tunnel, and known bug workarounds. Use whenever the user mentions installing NemoClaw, setting up OpenClaw, configuring an NVIDIA Spark or DGX for sandboxed agents, or troubleshooting NemoClaw deployment.893 docker-expertdavila7/claude-…emplatesDocker containerization expert with deep knowledge of multi-stage builds, image optimization, container security, Docker Compose orchestration, and production deployment patterns. Use PROACTIVELY for Dockerfile optimization, container issues, image size problems, security hardening, networking, and orchestration challenges.892 bash-scriptingmindrally/skillsBash scripting guidelines covering security, portability, error handling, and automation best practices for DevOps.885 infisical-user-setup-guideinfisical/ai-skillsInteractive setup guide for using Infisical as a secret management tool in your projects. Helps users integrate Infisical into local development (CLI), Docker containers (build-time and runtime secret injection), CI/CD pipelines (GitHub Actions, GitLab CI), Kubernetes (Operator + CRDs), and application code (all 9 SDKs: Node.js, Python, Go, Java, .NET, Ruby, PHP, Rust, C++). Also walks through choosing and configuring any of the 13 machine identity auth methods (Universal, Token, Kubernetes, GCP, AliCloud, AWS, Azure, TLS Cert, OCI, OIDC, JWT, LDAP, SPIFFE). Use this skill whenever someone asks about: using Infisical, injecting secrets, infisical run, infisical init, connecting their app to Infisical, Docker secrets, Kubernetes secrets operator, machine identity setup, SDK initialization, CI/CD secret injection, or 'how do I get my secrets into my app'. Not for the Kubernetes Operator CRDs (infisical-kubernetes-operator), rendering secrets to files (infisical-agent), pushing secrets outward (infisical-secret-syncs), or human SSO login (infisical-sso).881 aws-cost-operationszxkane/aws-skillsAWS cost optimization, monitoring, and operational excellence expert. Use when analyzing AWS bills, estimating costs, setting up CloudWatch alarms, querying logs, auditing CloudTrail activity, or assessing security posture. Essential when user mentions AWS costs, spending, billing, budget, pricing, CloudWatch, observability, monitoring, alerting, CloudTrail, audit, or wants to optimize AWS infrastructure costs and operational efficiency.873 server-managementsickn33/agentic…e-skillsServer management principles and decision-making. Process management, monitoring strategy, and scaling decisions. Teaches thinking, not commands.872 clawsec-suiteprompt-security/clawsecClawSec suite manager with embedded advisory-feed monitoring, cryptographic signature verification, approval-gated malicious-skill response, and guided setup for additional security skills.869 holyclaude-ai-workstationreason-machines/trending-skillsDeploy a full AI coding workstation with Claude Code, web UI, headless browser, and 5 AI CLIs in a single Docker container868 pricing-packaging-trackerdeanpeters/prod…r-skillsTrack competitor pricing and packaging as a diffable time series. Use when monitoring tiers, gates, limits, and price moves on a monthly or quarterly cadence.865 nvidia-skill-findernvidia/skillsUse for NVIDIA-related requests where an NVIDIA skill might help, even if the user did not ask for a skill. Trigger on NVIDIA products, hardware, software, SDKs, GPUs, Jetson/JetPack/L4T/BSP/SDK Manager/driver/flashing/setup, CUDA, NIM, NeMo, Omniverse/OpenUSD/SimReady, Isaac, RAPIDS/cuDF, cuPyNumeric, cuOpt, Dynamo, Holoscan, TensorRT, DeepStream, VSS, TAO, NGC/NVCF. Do not use for generic non-NVIDIA route, optimize, deploy, AI, video, data, or infrastructure tasks.862 provider-upgradepulumi/agent-skillsUpgrade any Pulumi provider to a newer version and reconcile the resulting diff. Use when users want to upgrade or update a provider (including editing package.json, requirements.txt, pyproject.toml, go.mod, or Pulumi.yaml to bump a provider SDK), check for breaking changes before or during an upgrade, fix resources that broke after a provider upgrade, or resolve unexpected replacements, creates, or deletes in a post-upgrade preview, including pulumi-gcp v10 errors such as a missing gcp.notebooks or gcp.iap.Brand type or a required secretDataWoVersion. Applies to all providers (aws, azure-native, gcp, kubernetes, aws-native, cloudflare, datadog, etc.) — not just Tier 1. Do NOT use for querying which stacks use what package versions; use skill `package-usage` for cross-stack audits. Do NOT use for general infrastructure tasks.860 dx-devops-conflict-resolveforcedotcom/sf-skillsUse this skill to diagnose and resolve what blocks a DevOps Center promotion of a work item's feature branch: Git merge conflicts and deployment failures. DevOps Center is Git-backed, so overlapping metadata changes surface as merge conflicts against the target stage branch, resolved with git (detect, resolve markers, commit, push). Deploy failures often cite a missing dependency a full promotion can fix when the component already exists on the branch. TRIGGER when the user wants to check a work item for conflicts before promoting, resolve merge conflicts or leftover conflict markers in metadata files (.xml, .object-meta.xml, .cls), reconcile a feature branch with its target stage branch, or diagnose why a DevOps Center promotion or deployment failed and whether a full promotion or a missing-dependency fix will resolve it. DO NOT TRIGGER for running the promotion itself (use dx-devops-promote), creating or updating work items (use dx-devops-work-item-manage), or deploying metadata directly to an org.859 netsuite-sdf-project-documentationoracle/netsuite…loud-sdkGenerate enterprise-grade documentation for NetSuite SDF projects. Analyze scripts, object XML files, `manifest.xml`, and SuiteQL queries to produce README.md, architecture diagrams (Mermaid/ASCII), deployment guides, and troubleshooting tables. Can integrate with post-deployment documentation workflows when automation (for example, hooks) is available.853 mcloud-localmedusajs/medusa…t-skillsExecute mcloud local build to reproduce a Cloud build on the local machine. Use when debugging a build-failed deployment without pushing to the tracked branch, iterating on a build fix, or testing build-variable changes locally. Requires Docker and must run inside the project's Git repo.851 service-catalog-template-deployforcedotcom/sf-skillsDeploys a chosen Unified Catalog Service Process template into a Salesforce org using the Salesforce CLI (sf), resolving the template by name against the live catalog and verifying by re-reading. Deterministic: it deploys only a template the user explicitly named, only when the name resolves to exactly one live template, and never guesses. Use when a business user asks to deploy, install, activate, set up, or provision a specific Unified Catalog or Service Process template they already identified — e.g. 'deploy the Reset Account Password template' or 'install the laptop request service process'. Triggers on: deploy a catalog template, install a Service Process template, set up the X template, activate the X service process. DO NOT TRIGGER when: the user is still searching or comparing templates and has not named one (use service-catalog-template-search), or the request concerns Data Cloud data kits, CRM Analytics, or App Framework templates rather than Unified Catalog Service Process templates.850