DevOps & Infra

harsh817 Bundle

Audit deployment and rollback readiness before production. Use when verifying build pipeline, staging deployment, environment configuration, migrations, smoke tests, release controls, rollback plan, and deploy safety.

ryankolean Bundle

Designs, captures, and deploys Claude Code subagents (markdown files in .claude/agents/ or ~/.claude/agents/). Use when the user asks to design, create, scope, or build a subagent, lock a working chat pattern as a subagent, deploy an existing subagent file, convert a skill or slash command into a subagent, or references the /agents command, .claude/agents/ directory, or subagent frontmatter. Not for AGENTS.md, CLAUDE.md, slash commands, hooks, or general skills (use prompt-to-skill for skills).

ryankolean Bundle

Expert architect-level review system that takes a software design document and systematically refines, reworks, and hardens it into a foolproof implementation blueprint with no blind spots. Runs a multi-phase pipeline: ingest, structural audit, technical deep dive, continuity check, gap analysis with autoheal, adversarial final review, and hardened output. Activates when the user says "review this design doc", "harden this design", "audit this spec", "review my architecture", "design doc review", or when the user provides a software design document and asks for feedback, critique, improvements, or validation. Also activate when the user uploads a technical spec, RFC, architecture doc, or system design and wants it improved, stress-tested, or made implementation-ready. Use this skill even when the user simply says "review this" and the input is clearly a software design document.

ryankolean Bundle

summit-video-pipeline

jorinyang Bundle

阿里云函数计算(FC)部署 Python 函数——通过 OpenAPI 创建服务/函数/HTTP触发器，含 ACS3-HMAC-SHA256 签名算法、WSGI handler 模板、常见坑位。触发：部署FC/函数计算/阿里云函数/Serverless部署/fcapp.run/CreateFunction。

jorinyang Bundle

题库系统全栈开发 — 从文档解析→数据库入库→Web SPA展示的完整管线。覆盖 docx/xlsx/md 题目提取、Supabase JSONB 存储、前端大屏展示、考试系统（登录/答题/评分/错题分析/后台管理）、LLM 主观题评分、OSS 部署。

jorinyang Bundle

GitHub skill-repo sync cron troubleshooting — class-level pitfalls encountered when running the daily `github-release-readme` pipeline against jorinyang/awesome-skills from a Windows cron host. Use when the sync shows unexpected sync candidates, deletes tracked files, double-counts or under-counts skills, or fails to rebase correctly. Each pitfall has a tested recipe. Captures class-level patterns; session-specific incidents belong in `references/execution-log-YYYY-MM-DD.md`.

jorinyang Skill

当 Agent 即将声称任何工作完成时必须触发——禁止无证据声明。覆盖：文件写入/OSS上传/飞书写入/技能创建/部署/测试通过/bug修复/PR提交。触发词：完成了/搞定/修好了/通过了/build成功/发布/deploy/提交/上传完成/文档已创建/部署完成/可以了/OK了/DONE

itamarzand88 Skill

<!-- source: prowler-pytest — https://raw.githubusercontent.com/prowler-cloud/prowler/master/skills/pytest/SKILL.md -->

itamarzand88 Skill

<!-- source: wshobson-ml-pipeline-workflow — https://raw.githubusercontent.com/wshobson/agents/main/plugins/machine-learning-ops/skills/ml-pipeline-workflow/SKILL.md -->

itamarzand88 Skill

<!-- source: wshobson-terraform-module-library — https://raw.githubusercontent.com/wshobson/agents/main/plugins/cloud-infrastructure/skills/terraform-module-library/SKILL.md -->

deankelly751 Skill

Design a CI/CD pipeline from scratch. Step-by-step procedure covering stage design, artifact management, environment promotion, and failure handling. Use when setting up CI/CD for new projects or improving existing pipelines.

deankelly751 Skill

Build a retrieval-augmented generation pipeline. Step-by-step procedure covering document ingestion, chunking, embedding, retrieval, and generation with citations. Use when implementing RAG systems, improving retrieval quality, or designing knowledge-augmented AI features.

deankelly751 Skill

Continuous integration and continuous delivery philosophy. Covers CI/CD principles, pipeline stages, deployment strategies, and the path from commit to production. Use when designing CI/CD pipelines or reviewing deployment processes.

deankelly751 Skill

Twelve-Factor App methodology for building cloud-native services. Covers all 12 factors with modern interpretations and practical decision guidance. Use when designing services for cloud deployment, containerization, or reviewing application portability.

deankelly751 Skill

Cloud architecture principles covering well-architected frameworks, multi-tier design, resilience patterns, and cost optimization. Provider-agnostic patterns for cloud-native systems. Use when designing cloud deployments, reviewing infrastructure architecture, or making cloud service decisions.

arjunprabhulal Skill

Establishes Git as the single source of truth for deployed state, with a pull-based controller reconciling the cluster to match a repo instead of humans or pipelines pushing changes via kubectl or helm. Use this whenever the user designs a deployment repo layout, asks how environments should be promoted, debates config repo vs app repo, wants rollback to mean "revert a commit," or decides what belongs in Git versus a secret store. For the controller itself use `argocd-operations`; for canary mechanics use `progressive-delivery`; for building images use `ci-pipelines`.

arjunprabhulal Skill

Covers caching and serving content at the edge — cache keys and TTL design, invalidation strategies, origin shielding, deciding what is safely cacheable, and moving static and dynamic content closer to users. Use this whenever the user is configuring a CDN, choosing cache-control headers, debugging stale or leaking cached content, setting up cache invalidation after a deploy, or deciding whether a response is safe to cache at all. For application-layer and database caching use `caching-strategies`, and for the origin's own load-balancing behavior use `load-balancing`.

arjunprabhulal Skill

Covers functions and managed compute where the platform enforces statelessness and bills per request — cold starts, event-driven design, concurrency limits, and when serverless does not fit. Use this whenever the user is writing a Lambda/Cloud Function/Azure Function, designing an event-driven pipeline, debugging cold-start latency, hitting concurrency throttling, or asking whether to move a workload to serverless. For managed-vs-self-run tradeoffs use `cloud-architecture`; for containers use `kubernetes-operations`; for scheduled batch use `scheduled-jobs`.

arjunprabhulal Skill

Covers running across cloud providers on purpose — portability vs managed services, the real operational cost of a second provider, avoiding accidental multi-cloud, and where the abstraction is worth it. Use this whenever the user is considering a second provider, evaluating a portability layer like Kubernetes across clouds, discovering their stack already spans providers without a plan, or asking whether to abstract away managed services. For the shape within one provider use `cloud-architecture`; for moving a workload use `cloud-migration`.

arjunprabhulal Bundle

Designs continuous integration pipelines that give a fast, honest merge signal — stage ordering, reproducibility, safe caching, and required checks that actually gate merges. Use this whenever the user is writing or debugging a CI workflow (GitHub Actions, GitLab CI, Jenkins), complaining that CI is slow or flaky, adding a required status check, or asking how to structure test stages. For what happens after merge use `continuous-delivery`; for making the underlying build itself faster use `build-optimization`; for securing the pipeline's runners and permissions use `pipeline-security`.

arjunprabhulal Skill

Matches compute, memory, and storage allocation to real measured usage instead of the guess made at launch time, sizing from percentiles rather than averages, and preferring autoscaling over a fixed size wherever demand varies. Use this whenever the user asks whether an instance or fleet is oversized, wants to cut provisioned capacity, is picking an instance type, or is choosing between fixed capacity and autoscaling. For turning savings into a prioritized plan use `cost-optimization`, and for the commitment discounts that should follow a sized fleet use `cloud-budgeting`.

arjunprabhulal Bundle

Covers enforcing infrastructure and cluster rules automatically, before a bad change ever reaches production — OPA, Sentinel, and Kyverno policies evaluated against the plan or admission request, and testing those policies like real code. Use this whenever the user is writing a Rego or Sentinel policy, adding an admission controller, wiring a policy check into a pipeline, or deciding whether a rule should be enforced or just audited. For the Terraform plan those policies gate use `infrastructure-as-code`, for Kubernetes-specific admission enforcement use `kubernetes-security`.

arjunprabhulal Skill

Decouples deploying code from releasing it to users through runtime feature flags — flag types, targeting rules, and the flag-debt cleanup discipline that keeps the flag system from becoming its own liability. Use this whenever the user wants to gate a feature behind a flag, is asking about kill switches, needs to roll out a change gradually to a subset of users, or has accumulated stale flags nobody remembers the purpose of. For strategies that use flags as part of a rollout mechanism use `deployment-strategies`; for the broader deploy-vs-release distinction use `continuous-delivery`.

arjunprabhulal Bundle

Covers changing schema or data shape without downtime using expand-then-contract — adding new structures alongside old, backfilling historical data in batches, dual-writing during the transition, verifying both old and new code paths, and keeping every step independently reversible. Use this whenever the user renames a column, changes a data type, moves data between tables, backfills a new field, or asks how to migrate without a maintenance window. For deploy sequencing of each phase use `deployment-strategies`, and for day-to-day database operations use `database-operations`.

arjunprabhulal Skill

Guides moving workloads to or between clouds using the 6 Rs, a phased cutover with a real rollback path, data sync, and avoiding a lift-and-shift that just relocates old problems. Use this whenever the user is planning a cloud migration, choosing between rehost/replatform/refactor, designing a cutover plan, migrating a database to a new provider, or asking why a "simple" migration is stalling. For the target shape use `cloud-architecture`; for sync mechanics use `data-migration`; for running providers long-term use `multi-cloud`.

arjunprabhulal Skill

Covers designing Terraform modules that are reusable and composable rather than copy-pasted or over-engineered — clean input/output interfaces, version pinning, and knowing when abstraction earns its complexity. Use this whenever the user is writing a new module, designing variables and outputs, deciding whether to nest modules, pinning a module source to a version, or debating whether shared logic should become a module yet. For state and plan discipline see `infrastructure-as-code`, for blocking bad module usage before apply use `policy-as-code`.

arjunprabhulal Skill

Covers the virtual network layer of a cloud deployment — VPCs, subnets, route tables, peering and transit, private endpoints, egress control, and hybrid or on-prem connectivity. Use this whenever the user is designing a VPC/VNet layout, planning subnet CIDR ranges, connecting two networks or a cloud to a data center, locking down egress, or debugging traffic that can't cross a network boundary. For DNS and load balancing use `dns-management`/`load-balancing`; for perimeter posture use `network-security`.

arjunprabhulal Skill

Forecasts cloud spend from trend and known upcoming changes, sets budgets and alerts that fire before an overrun becomes a surprise invoice, catches anomalies early, and turns raw spend into unit economics and showback/chargeback that leadership can act on. Use this whenever the user asks for a spend forecast, wants a budget alert or anomaly detector, is building a showback or chargeback report, or asks what a feature or customer costs to run. For cutting the spend a budget flags as too high use `cost-optimization`, and for the tags budgets are sliced by use `resource-tagging`.

arjunprabhulal Skill

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 autoscaling that flaps, or sizing a cluster autoscaler. For the requests/limits autoscaling depends on use `kubernetes-operations`; for cost tradeoffs use `cost-optimization`.

arjunprabhulal Skill

Covers packaging and templating Kubernetes manifests with Helm — chart structure, values design and environment overrides, releases and revisions, upgrade/rollback semantics, avoiding template sprawl, and choosing Helm versus Kustomize. Use this whenever the user is writing a Helm chart, designing a values.yaml hierarchy, debugging a failed `helm upgrade`, or deciding between Helm and Kustomize. For rollout mechanics on the cluster use `kubernetes-operations`; for GitOps delivery of charts use `gitops`.

arjunprabhulal Bundle

Secures the CI/CD pipeline itself as an attack surface — least-privilege runners, protecting secrets, preventing poisoned-pipeline execution, pinning third-party actions by SHA, and preferring OIDC over long-lived keys. Use this whenever the user configures CI runner permissions, stores secrets for a pipeline, reviews a workflow file for security issues, or sets up cloud credentials for a deploy job. For scanning produced artifacts use `image-scanning`; for the wider software supply chain use `supply-chain-security`.

arjunprabhulal Skill

Covers running stateful systems — databases, queues, search indexes — on Kubernetes, including StatefulSets and stable identity, durable storage, backup and failover built into the platform rather than bolted on, and the tradeoff between self-managing a stateful service and paying for a managed one. Use this whenever the user deploys a database or queue on Kubernetes, picks a StorageClass, debugs a pod that lost data on restart, or debates self-hosting versus managed. For volume mechanics use `kubernetes-storage`, and for restore discipline use `backup-and-restore`.

arjunprabhulal Skill

Builds a tag taxonomy for cost, ownership, and automation, enforces it at provision time so it never depends on discipline after the fact, and uses it to drive cost allocation, showback, and the hunt for untagged waste. Use this whenever the user is designing tag keys or naming conventions, asking who owns a resource or why a bill can't be attributed to a team, setting up cost allocation, or finding resources that escaped a tagging policy. For turning tags into dollar decisions use `cost-optimization`, and for the budgets tags feed into use `cloud-budgeting`.

arjunprabhulal Skill

Covers when and how to adopt a service mesh — mTLS between services, traffic shifting, retries/timeouts enforced at the mesh layer, near-free observability, and the real latency and complexity cost of running one. Use this whenever the user asks whether they need a service mesh, is configuring mTLS or traffic-splitting rules, or is debugging sidecar-added latency. For basic Service/Ingress routing use `kubernetes-networking`; for canary rollout mechanics use `progressive-delivery`.

arjunprabhulal Skill

Designs systems for the cloud's actual shape — regions and availability zones, managed vs self-run tradeoffs, statelessness, failure-domain isolation, and the cost and lock-in consequences of each choice. Use this whenever the user is choosing a region topology, deciding between a managed service and self-hosting, sketching a new system's compute and data layout, or asking why an architecture is expensive or fragile. For the virtual network layer use `cloud-networking`; for moving an existing workload use `cloud-migration`; for scoring an existing system use `well-architected-review`.