a2a-agent
Comprehensive A2A (Agent-to-Agent) development toolkit for building Python agents using a2a-sdk and FastAPI. Use when creating new A2A agents, implementing agent handlers, adding streaming capabilities, testing agents, or deploying agents. Triggers include "a2a agent", "agent-to-agent", "a2a-sdk", "create agent", "agent handler", "streaming agent", or when working with FastAPI-based agent architectures.
mcp-builder
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
microsoft-agent-framework
Expert guidance for building AI agents using Microsoft Agent Framework. Use when creating single agents, multi-agent workflows, chat applications, or integrating tools with LLMs. Covers ChatAgent, tools (@ai_function), orchestration patterns (GroupChat, Sequential, Concurrent, Handoff), memory/state management, observability, and declarative YAML agents. Supports Python and .NET.
write-skill
Use when creating a new skill or updating an existing skills that extends Claude's capabilities.
parallel-agent-spawner
Spawn and coordinate parallel agents for faster completion. Use when running parallel tasks, spawning subagents, coordinating concurrent work, or optimizing throughput.
telegram-web
Enables Claude to send messages, manage chats, and handle Telegram communications through the web interface
llm-app-architecture
Automatically applies when building LLM applications. Ensures proper async patterns for LLM calls, streaming responses, token management, retry logic, and error handling.
ralph-loop
Ralph Wiggum 自主循环模式。AI 持续执行任务直到全部完成，按用户确认的节奏汇报（默认整单完成后）。触发词：ralph、ralph-loop、/ralph-loop、自主循环、autonomous、直到完成、循环执行。
meta-router
Primary intent classifier and entry point for all skill routing. Routes requests through the unified 7-router hierarchy based on intent analysis. Triggers on any task requiring specialized skills or delegation.
autonomy-kickoff
Run the full autonomy bootstrap with minimal questions.
your-skill-name
Brief description of what this Skill does and when to use it. This field is critical for Claude to discover when to invoke your skill.
conductor
CRITICAL: Load and read ENTIRE skill content before ANY coding task. This skill defines orchestration rules and role-based delegation. DO NOT skip. DO NOT skim. Read the full SKILL.md content first. NEVER use built-in subagents - delegate via MCP roles defined in this skill. TRIGGERS: code, file, function, class, bug, error, fix, debug, refactor, implement, build, test, deploy, analyze, review, search, explore, find, plan, design, architecture, optimize, improve, migrate, security, performance
static-audio-generation
Generate and manage static audio files for Bob The Skull using ElevenLabs TTS. Use when creating greetings, startup messages, error responses, or test audio. Handles generation, naming, directory structure, and cross-repo syncing.
agent-builder-pydantic-ai
Build conversational AI agents using Pydantic AI + OpenRouter. Use when creating type-safe Python agents with tool calling, validation, and streaming.
mcp-builder
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services.
terminal-bench-strategies
Comprehensive strategies and patterns for maximizing Terminal-Bench 2.0 task completion. Contains battle-tested approaches from top-performing agents.
dev-prompt-engineering
Anthropic의 Best Practice를 기반으로 고성능 프롬프트를 설계하고 최적화합니다.
test-printing
Simple skill that prints a templated message with agf_id and prompt
