aws-architecture-reviewAWS Architecture Review process based on the Well-Architected six-pillar checklistaws-cost-optimizationAWS cost optimization with billing analysis, RI/SP evaluation, storage optimization, idle resource identification, and interactive cost reportsclient-weekly-reportClient weekly report workflow — summarize work by client with this week's progress and next week's plancompetitor-analysisCompetitive benchmarking workflow with cloud provider product mapping and comparison report templateshtml-presentationHTML slide creation workflow using reveal.js to generate interactive presentationssolution-documentTechnical solution document generation workflow with templates, architecture diagram integration, and document structure guidelines
リポジトリエクスプローラー
aws-samples · リポジトリと代表的な skills - 5ページ
176 件の GitHub リポジトリにある 1,166 件の収集済み skills をリポジトリ単位で表示します。
api-keysDual-mode API key storage — native file-based or AWS Secrets Manager. Store, retrieve, list, and delete API keys securely. Use manage_native for simple file storage, manage_secret for Secrets Manager with audit trail. Use retrieve to look up a key from either…eventbridge-cronSchedule recurring tasks using Amazon EventBridge Scheduler. Create, list, update, and delete cron schedules that fire at specified times and deliver results to the user's chat channel. Use when the user asks to schedule reminders, recurring tasks, or…s3-user-filesPer-user persistent file storage backed by AWS S3. Read, write, list, and delete text files isolated per user. Each user has a private S3 namespace — no cross-user data leakage. Use when the user asks to save, remember, or persist any information across…agentcore-browserBrowse web pages using a headless Chromium browser running inside the AgentCore container. Navigate to URLs, take screenshots, and interact with page elements (click, type, wait, scroll). Requires enable_browser=true in CDK config. Use when the user asks to…clawhub-manageInstall, uninstall, and list ClawHub community skills. Use when the user asks to install a new skill, remove an existing skill, or see what skills are available. Skills are installed globally in the container and available after OpenClaw restarts or on new…
spec-driven-presentation-makerGenerate PowerPoint presentations from JSON. Use when user wants to create slides, proposals, or presentation materials.sdpm-translate既存のスライドデッキ（sdpm デッキ）を別言語に翻訳した派生デッキを作るとき。 「このデッキを英語にして」「日本語版を作って」「translate this deck」などで起動。 元デッキは変更せず、隣に言語違いのデッキを生成する。 Translate an existing sdpm deck into another language as a derived deck. The source deck is left untouched.sdpm-specスライド / プレゼン資料 / PowerPoint を spec 駆動で、丁寧なヒアリングを通して作るとき。 要件・論理構造・デザインを対話で固め、各ステップでユーザー承認を取る。既存 PPTX の編集や ガイド付きメニューからのスタイル作成もこちら。「相談しながら資料を作りたい」で起動。 Dialogue-driven deck design — real hearing, structured requirements, approval at each step. Also the entry point…sdpm-style再利用できるスライドのスタイルガイド（HTML デザイントークン）を対話で作るとき。 「うちのブランドカラーでテンプレートを作りたい」「この見た目を使い回したい」で起動。 Create a reusable presentation style guide (HTML design tokens) through dialogue.sdpm-vibe手元の素材（URL・論文・議事録・PDF・貼り付けテキスト）から、ヒアリング最小・確認なしで スライド / プレゼン資料 / PowerPoint を一気に高速生成するとき。 「これをスライドにして」「このURLから資料作って」「ざっとパワポに」などで起動。 Fast, autonomous slide/deck/PowerPoint generation from source material the user already has. Minimal questions, no per-step…
moca-guideHow to use Moca itself — what this platform can do and how to ask for it. Covers building and running agents, scheduled/event triggers, long-term memory, Generative UI, the built-in tools, workspace file persistence, and model/reasoning selection. Activate…deep-modulesApply Ousterhout's "deep modules" principle (narrow interface, deep implementation) when designing or refactoring classes and modules. Use when adding new methods to a class, extracting shared logic, or reviewing repository / service code.doc-drift-detectionDetect semantic drift between documentation and source code. Goes beyond grep-based checks to find meaning-level discrepancies that mechanical tools miss.coding-styleDesign decisions, implicit rules, and anti-patterns for the Moca project. Covers coding conventions, project structure rationale, and testing guidelines.ui-designUI/UX design rules, Atomic Design placement, design tokens, component conventions, and responsive patterns for the Moca frontend. Use when creating or modifying React components, styling, or layout.
sales-report이커머스 매출 데이터를 표준 경영진 리포트 형식으로 작성하는 방법. 월간/주간 매출 리포트, 매출 분석 요약, 카테고리별 실적 보고를 요청받았을 때 사용한다.3a-guideUse when a developer asks about agent design best practices, patterns, or how to make a better agent. Provides curated guidance on architecture patterns, tool design, prompt engineering, error handling, and other quality aspects of agent development. Triggers…3a-planUse when a developer wants to plan an AI agent project before building it. Gathers requirements, defines architecture, and produces an execution roadmap. Triggers on: "plan my agent", "what should I build", "agent architecture", "requirements", "design my…3a-reviewUse when a developer is about to transition between project phases and needs a readiness check. Validates that prerequisites are met before moving from planning to implementation, implementation to deploy, or deploy to production. Triggers on: "am I ready to…3a-trackUse when a developer wants to check progress, understand where they are in their agent project, or figure out what to do next. Reads the roadmap and project state, then provides a status report with next steps. Triggers on: "where am I", "what's next",…
e2eThis skill should be used when the user asks to run E2E tests "in parallel", "with multiple agents", "並列で", "サブエージェントを立てて", or wants comprehensive E2E coverage split across concurrent agents. Fans out one `e2e-test` subagent per section group of…pre-push-reviewThis skill should be used when the user asks to "review before push", "pre-push check", "is this ready to push", or mentions push readiness. Runs code quality, security, test coverage, dead-code, AWS safety, and documentation checks.bump-versionThis skill should be used when the user asks to "bump version", "update version", "release patch/minor/major", or mentions semantic versioning. Increments the project version.run-testsThis skill should be used when the user asks to "run tests", "test my changes", "check if tests pass", or mentions testing changed files. Detects modified files and routes to the appropriate test suite.update-docsThis skill should be used when the user asks to "update docs", "sync documentation", "check if docs need updating", or mentions documentation consistency after code changes. Identifies and updates affected documentation.
bedrock-pricingUse when calculating Amazon Bedrock foundation model inference costs, comparing model pricing across tiers/regions, or estimating monthly spend for agentic workloads. Do NOT use for AgentCore infrastructure (load agentcore-pricing), RPM/TPM capacity (load…agent-business-valueUse when calculating business value, ROI, or cost justification for AI agents. Three dimensions: (1) Time Savings, (2) Customer Churn Reduction, (3) Sales Increase. Works standalone or alongside bedrock-pricing/agentcore-pricing estimates. Do NOT use for…agentcore-pricingUse when estimating Amazon Bedrock AgentCore infrastructure costs including Runtime, Gateway, Memory, BrowserTool, CodeInterpreter, or Evaluations. Do NOT use for model-only pricing (load bedrock-pricing), RPM/TPM capacity (load bedrock-capacity), or business…bedrock-capacityUse when checking RPM/TPM capacity limits, quota utilization, throttling risk, or capacity fit for Amazon Bedrock workloads. Also activate as a follow-up after bedrock-pricing estimates. Do NOT use for model pricing (load bedrock-pricing), AgentCore costs…bedrock-tier-advisorRecommends the right Bedrock service tier (Flex/Standard/Priority/Reserved/Batch) and inference variant (Global/Regional) based on workload characteristics. Activate when the user asks which tier to use, wants to optimize cost vs performance, or hasn't…
deploy-guideWalk through the full OpenCode Stack deployment step by stepdeploy-preflightValidate all prerequisites before deploying the OpenCode Stackjwt-debugDebug JWT authentication issues — decode tokens, check expiry, verify JWKS, inspect ALB rulestroubleshootDiagnose common OpenCode Stack deployment issuesverify-deploymentPost-deployment health checks and endpoint validation for the OpenCode Stack
the-care-cardCare, return, repair, and post-purchase language for moments where the shopper needs clear handling rather than more discovery.the-makers-shelfDeep knowledge of ceramics, textiles, and slow-craft home goods.the-packing-listTravel wardrobe curation — packable, natural, pieces that earn their place in the bag.the-proof-counterGovernance and inspectability language for "why this", "how do you know", memory proof, tool receipts, audit rows, Gateway traces, and source-grounded explanations.the-gift-tableExpert guidance on gift-worthy pieces — wrapping, milestones, considered giving.
aiplcAI Product Lifecycle pipeline — 从深度研究到交付的全自动四步流程。触发词: 'aiplc [主题]'、'AI-PLC [topic]'、'从研究到交付'。当用户想对一个主题进行完整的研究→PRD→计划→执行流程时激活。brainstormTransform a research markdown into a right-sized Product Requirements Document through forcing-question interrogation, premise pressure-testing, and adversarial self-review. Ingests research.md (or runs cold-start if none), classifies scope into Lightweight /…deepresearchDeep research orchestrator. Launches 5-8 parallel research agents across web, docs, code, and academic sources. Cross-validates findings via consensus matrix. Applies Socratic questioning to challenge assumptions and expose gaps. Generates structured markdown…executeConsumes a plan file produced by the plan skill and ships it to working artifacts. Dispatches independent units in parallel waves per the plan's dependency graph, runs a single consolidated Oracle review at the end, and updates the plan's Completion Status…planTransform a PRD (or conversation context) into a concise, DAG-shaped implementation plan with stable U-IDs, explicit dependencies, and declared parallel-execution waves. Consumes brainstorm-produced PRDs (R#/A#/F#/AE# IDs) or runs cold-start. Produces…
readme-generatorThis skill should be used when users want to create or improve README.md files for their projects. It generates professional documentation following the Deep Insight/Strands SDK style - comprehensive yet focused, with clear structure and practical examples.skill-creatorGuide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.system-prompt-writerThis skill should be used when writing or improving system prompts for AI agents, providing expert guidance based on Anthropic's context engineering principles.tool-creatorThis skill should be used when users want to create a new tool for the Strands SDK agent system. It supports creating both agent-as-a-tool (complex agents wrapped as tools) and regular tools (simple function-based tools). Use this skill when users request to…