메뉴
이 저장소의 skills
SkillsMP는 Sheshiyer/skill-clusters에서 722개의 skill을 수집했습니다. skill을 열어 소스와 세부 정보를 확인하세요.
Sheshiyer/skill-clusters수집된 skill 722개 중 40개를 표시합니다.
Image optimization analysis for SEO and Core Web Vitals — alt text, file sizes, WebP/AVIF formats, responsive srcset, lazy loading, fetchpriority, and CLS prevention. USE WHEN a user mentions image optimization, alt text, image SEO, image size, or an image…
원문 언어: 영어
Deep single-page SEO analysis covering on-page elements, content quality, technical meta tags, schema, images, and performance signals — returns a per-page score card and prioritized fixes. USE WHEN a user says analyze this page, check page SEO, or provides a…
원문 언어: 영어
Strategic SEO planning for new or existing websites — industry templates, competitive analysis, content strategy, KPI targets, and a phased implementation roadmap. USE WHEN a user wants an SEO plan, SEO strategy, content strategy, or SEO roadmap rather than a…
원문 언어: 영어
Analyze existing XML sitemaps or generate new ones with industry templates — validates format, URL limits, lastmod, canonical/noindex hygiene, and sitemap-index splitting. USE WHEN a user mentions sitemap, generate sitemap, sitemap issues, or XML sitemap.
원문 언어: 영어
Audit technical SEO across crawlability, indexability, security, URLs, mobile, Core Web Vitals, structured data, JavaScript rendering, and platform signals like robots.txt and AI-crawler access. USE WHEN a user wants a technical SEO review focused on…
원문 언어: 영어
Build 3D experiences for the web with Three.js, React Three Fiber, Spline, WebGL, and GLSL — product configurators, immersive sites, interactive scenes. USE WHEN adding 3D, a configurator, or WebGL depth to a web UI.
원문 언어: 영어
Find and fix WCAG 2.2 accessibility issues in code or live pages, in report mode (prioritized written audit, no edits) or fix mode (audit->edit->verify), preferring direct-CDP live-DOM auditing. USE WHEN an agent's UI output needs an accessibility audit or…
원문 언어: 영어
Diff a live page's accessibility violations against a baseline — by default compares uncommitted changes (stash-based), or pass --branch [<name>] to diff against a branch; reports only new violations introduced, violations fixed, and pre-existing count. USE…
원문 언어: 영어
Audit a live page for accessibility issues, locate each WCAG violation precisely, and return a selector-grounded fix worklist without editing. USE WHEN you want a full a11y audit of a running page or preview URL and a prioritized fix list (use accesslint-diff…
원문 언어: 영어
Test and benchmark LLM agents with behavioral contract testing, capability assessment, statistical/adversarial evaluation, regression pipelines, and production monitoring. USE WHEN you need to measure, benchmark, or regression-test an LLM agent's real-world…
원문 언어: 영어
Architect agent memory across short-term context, long-term vector stores, and CoALA-style cognitive layers, with LangMem patterns, vector-DB selection (Pinecone/Qdrant/Chroma), and chunking strategy. USE WHEN designing the memory architecture and…
원문 언어: 영어
Design agent tools from JSON Schema to error handling and MCP exposure, writing LLM-legible descriptions, validation, and token-efficient interfaces that avoid silent failures. USE WHEN defining or hardening the tools/functions an LLM agent can call.
원문 언어: 영어
Design and build autonomous AI agents: ReAct and plan-and-execute loops, tool registries, hierarchical memory, supervisor patterns, and checkpoint recovery. USE WHEN architecting an autonomous agent's control loop, tools, and memory from scratch.
원문 언어: 영어
Design spec with 98 rules for building CLI tools that AI agents can safely use — structured JSON output, error handling, input contracts, safety guardrails, exit codes, and agent self-description. USE WHEN building, retrofitting, or auditing a command-line…
원문 언어: 영어
Production-grade LLM product engineering — RAG architecture, prompt engineering that scales, structured-output validation, latency/streaming UX, and cost control. USE WHEN building an AI-powered product feature and deciding how to integrate an LLM so it…
원문 언어: 영어
Turn an AI API (OpenAI, Anthropic, etc.) into a focused, defensible, paid product — prompt engineering for products, usage metering, rate limiting, model selection, and output quality control. USE WHEN building or pricing an AI-wrapper business and you need…
원문 언어: 영어
Implement Algolia search: indexing strategies, React InstantSearch hooks/widgets, faceting, and relevance tuning. USE WHEN adding site/app search, type-ahead, faceted filtering, or tuning Algolia relevance.
원문 언어: 영어
Generate user-facing App Store release notes from git history since the last tag, grouped into New/Improved/Fixed with internal commits filtered out. USE WHEN drafting What's New text, release notes, or changelog bullets for an App Store / TestFlight…
원문 언어: 영어
Core Python package for astronomy: coordinate transforms, units/quantities, FITS I/O, WCS, cosmology, precise time handling, and table operations. USE WHEN a research task involves astronomical data, celestial coordinates, FITS files, unit conversions, or…
원문 언어: 영어
Build reliable autonomous agents that decompose goals, plan, execute tools, and self-correct, focusing on bounding failure probability across multi-step runs. USE WHEN making an unattended autonomous agent reliable rather than merely capable.
원문 언어: 영어
Audit and rewrite prose to strip 21 categories of AI writing patterns using a 43-entry replacement table, then re-audit the rewrite. USE WHEN asked to remove AI-isms, make text sound less like AI, or polish a draft for AI tells before publishing.
원문 언어: 영어
Production-ready serverless on AWS — Lambda, API Gateway, DynamoDB, SQS/SNS event-driven patterns, SAM/CDK deployment, and cold-start optimization. USE WHEN building, deploying, or tuning AWS serverless apps (Lambda handlers, event pipelines, DynamoDB access,…
원문 언어: 영어
AI-powered E2E web testing — declarative YAML scenarios run via Playwright with visual matching (OpenCV + OCR) and framework auto-detection. USE WHEN you need AI-driven browser end-to-end tests without brittle DOM selectors.
원문 언어: 영어
Azure Functions patterns — isolated worker model, Durable Functions orchestration, cold-start optimization, production hardening across .NET, Python, and Node.js. USE WHEN building, orchestrating, or optimizing Azure Functions apps (durable workflows,…
원문 언어: 영어
Build cross-browser extensions (Chrome/Firefox) — Manifest V3 architecture, content scripts, service workers, popup UIs, monetization, and Web Store publishing. USE WHEN building, structuring, or publishing a browser extension.
원문 언어: 영어
BullMQ expert for Redis-backed job queues, background processing, retries/backoff, dead-letter queues, concurrency limits, and reliable async execution in Node.js/TypeScript. USE WHEN an agent or backend needs durable background jobs or an async work queue on…
원문 언어: 영어
Route a business-content task to the right skill among 10 specialists — voice profiling, long-form articles, multi-platform content systems, SEO, market/competitor research, launch campaigns, investor decks/memos, investor outreach, lead intelligence, and…
원문 언어: 영어
Personal health-assistant patterns: medical-information analysis, symptom tracking, and wellness guidance for a patient/consumer-facing companion app. USE WHEN building a personal health assistant, logging or trending symptoms, or surfacing general wellness…
원문 언어: 영어
Build apps with the Claude API / Anthropic SDKs (TypeScript, Python, Agent SDK), choosing the right surface and language-specific patterns; not for OpenAI or other providers. USE WHEN code imports anthropic/@anthropic-ai/sdk/claude_agent_sdk or a user asks to…
원문 언어: 영어
Apply Robert C. Martin (Uncle Bob) Clean Code principles — meaningful names, small single-purpose functions, comment discipline, error handling, and code-smell heuristics — to transform code that works into code that is clean. USE WHEN reviewing a PR for…
원문 언어: 영어
Implement Clerk authentication — App Router setup, middleware-protected routes, organizations, webhooks, and user sync to your database. USE WHEN adding Clerk auth, org/multi-tenant, or user-sync webhooks to a Next.js app.
원문 언어: 영어
React composition patterns for flexible, maintainable components: compound components, lifting state, explicit variants, children over render props, and avoiding boolean-prop proliferation. USE WHEN refactoring React components that are drowning in boolean…
원문 언어: 영어
Build agents that operate a computer like a human, viewing screens and clicking/typing, across Anthropic Computer Use, OpenAI Operator/CUA, and open-source alternatives. USE WHEN building an agent that drives a GUI via screenshots and mouse/keyboard control.
원문 언어: 영어
Strategies for managing LLM context windows including summarization, trimming, routing, token counting, prioritization, and avoiding context rot. USE WHEN an agent run is overflowing its context window or degrading from context rot.
원문 언어: 영어
Persistent conversation memory with tiered short/long-term stores, entity memory, per-user isolation, importance filtering, and memory-aware prompting. USE WHEN an assistant must remember users across sessions without leaking memory between them.
원문 언어: 영어
Convex reactive-backend expert covering schema design, TypeScript queries/mutations/actions, real-time subscriptions, auth, file storage, scheduling, cron, and deployment. USE WHEN building or debugging an app on the Convex reactive backend.
원문 언어: 영어
CrewAI expert for role-based multi-agent systems: agents with roles/goals, task definitions, sequential/hierarchical/parallel crews, memory, and flows. USE WHEN orchestrating a multi-agent crew with the CrewAI framework.
원문 언어: 영어
Extract clean markdown content from web pages using the Defuddle CLI, removing nav/ads/clutter to save tokens. USE WHEN a user gives a URL to read, summarize, or analyze (docs, articles, blog posts, standard pages) and you want lean content instead of a raw…
원문 언어: 영어
Deploy applications and websites to Vercel, preferring git-push preview deploys and proper project linking. USE WHEN the user says deploy my app, push this live, give me a link, or make a preview deployment.
원문 언어: 영어
Creates devcontainers with the agent CLI, language-specific tooling (Python/Node/Rust/Go), and persistent volumes. USE WHEN adding devcontainer support to a project, setting up isolated development environments, or configuring sandboxed coding-agent…
원문 언어: 영어