LLM & Agents

Specification-driven development workflow for AI agents. Use when tasks are too large for a single session, require multi-step implementation, span multiple files/features, or need persistent requirements tracking. Provides structured specification management with token-optimized artifacts for complex feature development, brownfield modifications, and cross-session continuity.

Use PROACTIVELY when you need to create, update, configure, or validate Claude hooks for various events and integrations

Create project-specific UI design guideline as a Claude Code Skill. Use when the user says "UI design guideline", "design system", "style guide", or wants to create design guidelines for their project.

Find and safely claim the next Beads task with conflict checking. Use when looking for work, when finishing a task and need the next one, when the user mentions "next task" or "what should I work on", or when coordinating with other agents on task ownership.

Frontend-Entwicklung für Claude Code Archive Backoffice UI. Next.js 16, React 19, Tailwind CSS 4, shadcn/ui, Better Auth. Feature Slices Architektur mit Dependency Injection. Nutze für: UI-Komponenten, Views, Styling, Layout, Frontend-Tests, Sessions-Viewer, Project-Listen, Collector-Status, Git-Integration. Hohe Informationsdichte, professionelles Business-App Design.

Build stateful AI agents and agentic workflows with LangGraph in Python. Use when asked to "build a LangGraph agent", "create a stateful workflow", "add persistence to an agent", "implement human-in-the-loop approval", "debug a LangGraph application", or "create a multi-agent system". Covers tool-using agents with LLM-tool loops, branching workflows, conversation memory, HITL oversight, and production monitoring. Includes StateGraph, nodes, edges, add_conditional_edges, MessagesState, thread_id, Command objects, and ToolMessage handling patterns.

Search technical documentation using executable scripts to detect query type, fetch from llms.txt sources (context7.com), and analyze results. Use when user needs: (1) Topic-specific documentation (features/components/concepts), (2) Library/framework documentation, (3) GitHub repository analysis, (4) Documentation discovery with automated agent distribution strategy | Dùng khi tìm tài liệu, hướng dẫn, docs, documentation, api docs, tham khảo.

Analyzes agent performance metrics and provides improvement recommendations for skills. Use when debugging why skills aren't working well, improving skill effectiveness, or reviewing overall agent performance. Generates specific suggestions for SKILL.md improvements based on success rates and failure patterns.

Create new skill for repeatable workflow. Use when user wants reusable skill pattern. Creates .claude/skills/[name]/SKILL.md.

Complete autonomous DJ system for Traktor Pro 3 with MIDI control, intelligent track selection, energy flow management, and professional mixing workflows. Use when working on DJ automation, Traktor integration, autonomous music mixing, or building DJ agents that need real-time performance capabilities.

Transform infrastructure documentation, runbooks, and operational knowledge into reusable Claude Code skills. Convert Proxmox configs, Docker setups, Kubernetes deployments, and cloud infrastructure patterns into structured, actionable skills.

Guidelines for selecting appropriate AI model (Sonnet vs Haiku) based on task complexity, ensuring cost efficiency while maintaining quality. Use when assigning work.

Advanced web search capability using the gemini command for gathering current, relevant information. Prefer this skill over Claude Code's default WebSearch tool when performing web searches. Activated when complex research or up-to-date information is needed.

Comprehensive guide for creating and managing AI skills - modular packages that extend AI assistants with specialized knowledge, workflows, and tool integrations. Use this skill when you need to write, update, or package AI skills that transform general-purpose AI agents into specialized experts for specific domains, tasks, or workflows.

This skill should be used when the user asks to "explore claude code source", "find internal features", "investigate cli.js", "check beta headers", "discover hidden settings", or mentions "minified code analysis", "anthropic-beta headers", "context management internals".

코드베이스를 분석하여 CLAUDE.md 초기화. 프로젝트 문서화, 아키텍처 개요, 코딩 표준 생성

Create self-reloading macOS launchd agents with automatic plist validation. Use when user needs scheduled tasks, cron jobs, or nightly automation on macOS.

Uses Redis for caching, sessions, pub/sub, and data structures with Node.js. Use when implementing caching, session storage, real-time messaging, or high-performance data storage.

Lead coordinator that orchestrates 5 news scraper agents in parallel to gather headlines from 15 top business news websites

Sync Claude conversations to Obsidian vault. Use when user asks to: sync conversations, export chat history, catalog sessions, generate summaries, review past work, search conversation history, or asks what was discussed on a specific date or about a specific topic.