LLM & Agents

LangGraph low-level agent orchestration. Build stateful, durable agents with graphs, persistence, streaming, and human-in-the-loop.

Use when creating, improving, or troubleshooting Claude Code subagents. Expert guidance on agent design, system prompts, tool access, model selection, and best practices for building specialized AI assistants.

byobu/tmux 터미널 간 Claude 세션 상태 공유 및 양방향 통신 스킬. 여러 터미널에서 실행 중인 Claude 세션의 상태를 모니터링하고 메시지를 주고받을 수 있습니다. "byobu", "터미널 상태", "세션 공유", "터미널 통신", "멀티 터미널" 키워드로 활성화.

Write tests using Vitest and Testing Library. Use when asked to write tests, add test coverage, or test a component/function.

Expert guidance for building LLM applications with LangChain framework - chains, prompts, memory, retrievers, and integrations.

Expert Unreal Engine 5 developer and technical artist for complete game development via agentic coding. Enables AI-driven control of Unreal Editor through MCP, Python scripting, Blueprints, and C++ for level design, asset management, gameplay programming, and visual development.

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

Creates and validates Claude Code skills with proper format. Use when creating new skills, writing SKILL.md files, validating existing skills, or when user mentions "skill format", "create skill", "skill template", or "validate skill".

This skill should be used when the user asks about "multi-agent coordination", "agent spawning", "parallel agents", "agent communication", "team coordination", "sub-agent management", "agent limits", "agent layers", or needs guidance on coordinating multiple agents for complex tasks.

This skill should be used when the user asks to "find something to write about", "brainstorm topics", "what should I write about", "find writing ideas", "gather sources for", "pull references on", or needs help discovering topics from their reading highlights. Leverages Readwise MCP to surface patterns and gather references.

Allra 백엔드 테스트 작성 표준. Use when writing test code, choosing test helpers, generating test data with Fixture Monkey, or verifying test coverage.

Specialized agent for BarqNet client application development across Desktop (Electron/TypeScript), iOS (Swift), and Android (Kotlin). Handles UI/UX implementation, OpenVPN integration, platform-specific features, secure storage, and native API usage. Use when developing client-side features, fixing UI bugs, or implementing platform-specific functionality.

Analyzes subagent usage patterns to determine when to delegate tasks, evaluates and improves existing subagent configurations, creates new custom subagents when needed, and identifies outdated subagent information. Use when deciding if a task should use a subagent, analyzing existing agents for effectiveness, recommending agent improvements, creating specialized agents for recurring patterns, or managing agent configurations. Optimizes Claude's subagent usage for performance, context preservation, and task specialization.

Analyze recent changes and add test coverage for HEAD commit

Check comprehensive platform health including ArgoCD apps, pods, services, certificates, and resources across the Kagenti platform

Submit complex code questions to OpenAI GPT-5 Pro for deep analysis when you have 20 minutes. Use when the user asks for architectural analysis, comprehensive code review, debugging complex issues, or requests expert analysis of their codebase that requires extended reasoning. Automatically handles file selection, code packing with Repomix, cost estimation, long-running API requests, and result presentation.

Generate a skeleton template for a new Claude Code Skill.

This skill should be used for periodic review of all Claude Code skills. Triggers include "audit my skills", "which skills need work", "skill inventory", or when starting a skill improvement session. Identifies staleness, gaps, redundancy, and prioritizes improvements.

Codex CLI를 sub-agent로 활용하여 실제 구현 작업을 수행합니다. Claude가 작업을 분석/분해하고, Codex가 코드 작성/수정/리팩토링을 담당합니다. "Codex로 구현", "구현해줘", "코드 작성해줘", "병렬 구현" 요청 시 활성화됩니다.

A comprehensive skill that helps Claude review, analyze, and generate Ruby on Rails code according to established best practices.