LLM & Agents

Analyzes codebases by impact scope (low/medium/high), generates refactoring checklists, supports interactive review, executes refactors with QA, and produces detailed optimization reports.

Create API agents that wrap external HTTP services (n8n, LangGraph, CrewAI, OpenAI endpoints). Configure request/response transforms, webhook status tracking, A2A protocol compliance. CRITICAL: Request transforms use template variables ({{userMessage}}, {{conversationId}}, etc.). Response transforms use field extraction. Status webhook URL must read from environment variables.

Generate and edit images using AI models (Gemini, OpenAI, BFL FLUX, or Poe). Use this skill when you need to create images from text prompts, edit existing images, or generate variations. Supports multiple providers with automatic API key detection from environment variables.

多 Agent 協作討論模式。當用戶說「party mode」、「bmad party mode」、「開始/啟動 party mode」、「召喚團隊」、「召喚專家」、「團隊討論」、「專家會議」時啟動。適用於技術架構討論、產品策略規劃、專案腦力激盪等需要多角度專家觀點的場景。

Coordinate multi-step workflows involving multiple agents, commands, and validation points. Use when managing complex development workflows that require coordination between different specialized agents.

Use this skill when you need to orchestrate specialized Claude agents in separate sessions. Perfect for long-running tasks, specialized agents with MCP tools, and resumable workflows.

Single source of truth for MetaSaver agent selection and subagent_type mapping

AI video generation with Sora, Veo, and Kling. Use when creating Reels, video prompts, or handling video generation workflows.

A comprehensive skill designed for researching on complex diagnosis problems.

Integrates Anthropic Claude API for chat, tool use, and vision capabilities. Use when building AI-powered features with Claude models including streaming, extended thinking, and multi-turn conversations.

Skill for implementing frontend features following project conventions. Use when writing Vue components, creating UI, or refactoring frontend code. Loads all frontend rules from .claude/rules/frontend/.

Apply chronicle event effects to world state. Use when events need to modify entities (battles change force strength, treaties change relationships, deaths remove agents). This is the core simulation engine.

This skill provides comprehensive knowledge for building applications with Cloudflare Sandboxes SDK, which enables secure, isolated code execution in full Linux containers at the edge. It should be used when executing untrusted code, running Python/Node.js scripts, performing git operations, building AI code execution systems, creating interactive development environments, or implementing CI/CD workflows that require full OS capabilities.Use when: Setting up Cloudflare Sandboxes, executing Python/Node.js code safely, managing stateful development environments, implementing AI code interpreters, running shell commands in isolation, handling git repositories programmatically, building chat-based coding agents, creating temporary build environments, processing files with system tools (ffmpeg, imagemagick, etc.), or when encountering issues with container lifecycle, session management, or state persistence.Keywords: cloudflare sandbox, container execution, code execution, isolated environment, durable objects

Apply instruction-engineering to all subagent prompts. Use BEFORE invoking the Task tool, spawning agents, or dispatching parallel workers. Ensures subagents receive persona-driven, research-backed prompts that maximize compliance and output quality. Triggers on: "use a subagent", "spawn agent", "dispatch", "Task tool", parallel agent work, or any multi-agent orchestration.

Generate high-quality synthetic datasets using statistical samplers and Claude's native LLM capabilities. Use when users ask to create synthetic data, generate datasets, create fake/mock data, generate test data, training data, or any data generation task. Supports CSV, JSON, JSONL, Parquet output. Adapted from NVIDIA NeMo DataDesigner (Apache 2.0).

Implement real-time collaborative drawing canvas with brush tools, color picker, undo/clear functionality, and canvas-to-image export. Use when building drawing UI, handling stroke synchronization, managing drawing state, or uploading drawings to storage.

SQLAlchemy repository pattern implementation for LoRAIro database operations with type-safe transactions, session management, and ORM best practices

Anthropic 베스트 프랙티스를 따라 Claude Code 스킬을 생성하고 관리합니다. 새로운 스킬을 만들거나 skill-rules.json을 수정하고, 트리거 패턴을 이해하며, 훅을 다루고, 스킬 활성화를 디버깅하거나, 점진적 공개를 구현할 때 사용하세요. 스킬 구조, YAML 프론트매터, 트리거 타입(키워드, 의도 패턴, 파일 경로, 콘텐츠 패턴), 강제 수준(블록, 제안, 경고), 훅 메커니즘(UserPromptSubmit, PreToolUse), 세션 추적, 500줄 규칙을 다룹니다.

Use when updating Claude Dev Kit components - checks registry for new versions, shows changelogs, updates components individually or all at once, supports rollback

Build AI-powered features in Directus: chat interfaces, content generation, smart suggestions, and copilot functionality