Discover AI Agent Skills

Designing and auditing repository structure for humans and LLM agents: layouts, monorepos, docs/tests/scripts, progressive disclosure, prompt-cache topology, and safe migrations.

Build production LLM streaming UIs with Server-Sent Events, real-time token display, cancellation, error recovery. Handles OpenAI/Anthropic/Claude streaming APIs. Use for chatbots, AI assistants, real-time text generation. Activate on "LLM streaming", "SSE", "token stream", "chat UI", "real-time AI". NOT for batch processing, non-streaming APIs, or WebSocket bidirectional chat.

Prepare for job interviews across formats—behavioral, technical, portfolio reviews, and negotiations. STAR/CAR frameworks for stories, common question patterns, research strategies, and follow-up templates. Triggers on interview prep, job interviews, behavioral questions, or salary negotiation requests.

Help users design their marketing organizational structure and select the foundational technology stack needed to drive scalable growth.

Transform podcast transcripts into multiple content assets—blog posts, social snippets, newsletters, and SEO-optimized landing pages—using systematic repurposing workflows. Use when: Maximizing ROI from podcast episodes; Creating blog content from audio/video; Generating social media posts from long-form content; Building newsletter content from transcripts; Extracting quotes and highlights for promotion

Generate detailed content briefs for blog posts with target keywords, content outlines, competitive analysis, recommended statistics, image and chart suggestions, word count targets, internal linking architecture, template recommendations (12 types), TL;DR drafts, evidence-backed section planning, information gain prompts, and multi-channel distribution plans. Briefs are optimized for Google rankings and AI citation visibility as part of SEO. Use when user says "content brief", "blog brief", "write brief", "SEO brief", "article brief", or "content requirements".

Mid-session save point: create a descriptive git commit and a brief handoff note, then keep working. Use before a risky change or refactor, when switching tasks, or to bank progress without ending the session. Triggers on: /checkpoint, "checkpoint", "save progress", "commit and handoff", "save state", "pause here", "before a risky change". For the full end-of-session ritual with learning extraction, use /wrap-up instead.

REST/GraphQL API design patterns - resource naming, HTTP methods, error handling, pagination, versioning. Use when: design API, REST endpoints, GraphQL schema, error responses, pagination, rate limiting, API documentation.

测试驱动开发。适用于用户想用先写测试的方式构建功能或修复缺陷、提到 “red-green-refactor”，或需要集成测试时。

When the user wants to build or improve a sales bot's ability to detect and adapt to prospect communication preferences across SMS, email, phone, and chat. Also use when the user mentions "channel preference," "communication preference," "preferred channel," "omnichannel," or "channel switching."

Scan GDDs against the entity registry for cross-document conflicts. Grep-first approach targets conflicting sections, different stats.

Builds generative AI applications on Amazon Bedrock. Covers model invocation (Converse API, InvokeModel), RAG with Knowledge Bases, Bedrock Agents, Guardrails, and AgentCore (including the Harness managed agent loop). Use when invoking models, setting up Knowledge Bases, creating agents, applying guardrails, deploying to AgentCore, migrating/porting/converting a Bedrock Agent (including inline agents) to an AgentCore Harness, troubleshooting Bedrock errors (ThrottlingException, AccessDeniedException), or choosing models (Claude, Llama, Nova, Titan). ALSO USE for prompt caching, quota health checks and throttling diagnosis, cost attribution, migrating between Claude model generations, chunking strategies, API selection (Converse vs InvokeModel), and model selection. Also covers AgentCore Payments setup (x402, microtransactions, Payment Manager, Coinbase CDP, Stripe Privy, 402 Payment Required, paid endpoint). NOT for custom model training, Rekognition, or Comprehend.

Use when designing or building modern web applications in JavaScript/TypeScript (SPA/SSR/SSG/ISR/RSC): architecture trade-offs, state/data patterns, performance, testing, delivery, and team scaling.

设计、优化和测试AI模型提示词，使用系统化的提示工程技术，包括少样本学习、思维链和结构化输出。

Use for authored-source correspondence corpus maintenance: verify the vendored CIVIL/EVIL snapshot for drift against upstream, re-harvest or re-pin it, run the offline benchmark, and decide when to advance the pin.

Research and investigate business prospects and leads. Gathers company information, contact details, and qualification data for sales.

Reference for writing and editing skills well — the vocabulary and principles that make a skill predictable.

Evaluate demand-forecast quality honestly - WMAPE, bias and Forecast Value Added against a naive benchmark over a rolling-origin backtest. Use when the user mentions forecast accuracy, MAPE, demand planning performance, tahmin doğruluğu, talep tahmini, or asks whether a forecasting process or tool is worth it. Differentiator - judges the process (value added vs doing nothing), not just the model.