LLM 與 Agent

Builds AI agents using OpenAI Agents SDK with async/await patterns and multi-agent orchestration.Use when creating tutoring agents, building agent handoffs, implementing tool-calling agents, or orchestrating multiple specialists.Covers Agent class, Runner patterns, function tools, guardrails, and streaming responses.NOT when using raw OpenAI API without SDK or other agent frameworks like LangChain.

Patterns for building AI-powered chatbot backends using OpenAI Agents SDK with MCP (Model Context Protocol) server integration in FastAPI applications. Supports both standalone MCP servers and function tools.

Design, discover, and refactor multi-command workflows for Claude Code

Guides writing Vitest tests for Chrome extension code. Use when creating tests, improving coverage, or testing Vue components, content scripts, or Chrome APIs.

Comprehensive guide for creating effective Claude Code skills with security best practices, CLI-specific features, and structural patterns. Use when creating skills, needing security guidance, understanding skill architecture, or learning best practices.

Grades and improves CLAUDE.md (Claude Code) and AGENTS.md (Codex/OpenCode) configuration files. Use when asked to grade, score, evaluate, audit, review, improve, fix, optimize, or refactor agent config files. Triggers on 'grade my CLAUDE.md', 'score my AGENTS.md', 'is my CLAUDE.md too big', 'improve my agent config', 'fix my CLAUDE.md', 'optimize context usage', 'reduce tokens in CLAUDE.md', or 'audit my config files'. Automatically grades both files if present, generates improvement plan, and implements changes on approval.

Guide for discovering and using @j0kz MCP tools efficiently through search, filtering, and progressive exploration. Use when exploring available capabilities, finding the right tool for a task, or ...

Expert in implementing Microsoft 365 Copilot agents with TypeSpec code (use AFTER project is created). Use when (1) Writing or editing .tsp files in an existing project, (2) Implementing agent instructions and behavior in TypeSpec, (3) Adding or configuring capabilities (WebSearch, OneDrive, etc.) with proper scoping, (4) Creating API plugin actions with @actions decorator, (5) Defining TypeSpec models and enums, (6) Adding @doc decorators and documentation, (7) Compiling TypeSpec with npm run compile, (8) Fixing TypeSpec compilation errors, (9) Working with @agent, @instructions, @conversationStarter decorators, or (10) Implementing TypeSpec syntax for @microsoft/typespec-m365-copilot library. This skill handles TYPESPEC CODE, not project scaffolding.

Manages structured content with Sanity headless CMS using GROQ queries and real-time collaboration. Use when building content-driven sites with customizable schemas, live previews, and flexible data modeling.

Helps IDE agents navigate and operate the MOVA Skills Lab: read the registry, plan skill runs and manage episodes.

Write and run JavaScript/TypeScript tests using Vitest or Jest with mocking, component testing, and coverage. Use when writing JS/TS tests, testing Svelte/React components, or setting up test configuration.

Use when designing tests, implementing test cases, running tests, analyzing test coverage, or fixing test failures. Apply when user mentions testing, test cases, TDD, coverage, or asks to run tests. Use proactively after implementation is complete to ensure adequate test coverage.

Parallel development using git worktrees with Claude Code subagents. Use when spawning multiple async Claude agents across isolated worktrees, running parallel implementations, coordinating multi-agent workflows, leveraging LLM non-determinism for multiple solutions, or running background agents that signal completion. Triggers on "parallel agents", "worktrees", "subagents", "async Claude", "spawn agents", "parallel development", "multi-agent workflow", "background agents", "agent coordination", "parallel tasks".

Comprehensive draft story review with parallel specialist sub-agents. Spawns PM, UX, and SM agents to review stories from product, design, and implementation readiness perspectives before development begins.

This skill should be used when creating extensions for Claude Code or OpenCode, including plugins, commands, agents, skills, and custom tools. Covers both platforms with format specifications, best practices, and the ferg-engineering build system.

PRD 기반 프로젝트 부트스트래핑 & 스킬 관리 시스템.[Mode 1 - 새 프로젝트] "PRD 만들어줘", "새 프로젝트", "프로젝트 시작"[Mode 2 - 스킬 추가] "스킬 추가해줘", "frontend 스킬 만들어줘", "backend 스킬 적용"[Mode 2 - 디자인 참조] "이 사이트처럼 만들어줘", "이 디자인 참고해줘" + URL[Mode 2 - 동적 스킬] "AI 스킬 추가해줘", "이 패턴 스킬로 만들어줘" (템플릿 없어도 생성)[Mode 3 - 동기화] "PRD 변경사항 반영해줘", "스킬 동기화"[Mode 4 - 코드 역분석] "PRD 만들어줘" + 기존 코드 존재, "코드 분석해서 PRD 만들어줘"[Mode 5 - 문서 검증] "프로젝트 분석해줘" + PRD/CLAUDE.md 존재, "문서 검증해줘"

Creates a detailed technical design document from a product opportunity through collaborative workshopping. Use this when the user wants to create a technical spec, design document, or needs to plan a new feature or project through an interactive process with discovery, alternatives exploration, and iterative feedback before finalizing.

Searching internet for technical documentation using llms.txt standard, GitHub repositories via Repomix, and parallel exploration. Use when user needs: (1) Latest documentation for libraries/frameworks, (2) Documentation in llms.txt format, (3) GitHub repository analysis, (4) Documentation without direct llms.txt support, (5) Multiple documentation sources in parallel

Use when creating prompts for Claude or other LLMs, need help structuring prompts effectively, want to improve prompt quality, or need templates for simple or complex prompting tasks. Provides Anthropic's official 10-component framework, best practices for Claude 4.x models, and ready-to-use templates.

Add MCP servers to Claude Code configuration at user level (~/.claude). Supports stdio, HTTP, and SSE transports with environment variable prompting. Use when "add mcp server", "install mcp", "configure mcp server", "new mcp", or "setup mcp server".