Agent skill

custom-ai-agents

Stars 5
Forks 0

Install this agent skill to your Project

npx add-skill https://github.com/Gaku52/claude-code-skills/tree/main/07-ai/custom-ai-agents

SKILL.md

日本語版

Custom AI Agents

AI agents grant LLMs the ability to act autonomously. This skill covers everything about AI agent development — from agent architecture, tool integration, and multi-agent systems to the MCP protocol and production deployment.

Target Audience

  • Engineers designing and implementing AI agents
  • Those looking to build more advanced LLM applications
  • Those interested in multi-agent systems

Prerequisites

  • Experience using LLM APIs
  • Hands-on experience with TypeScript or Python
  • Web development fundamentals

Learning Guide

00-fundamentals — Agent Fundamentals

# File Description

01-architecture — Architecture

# File Description

02-tools — Tools and Frameworks

# File Description

03-advanced — Advanced Topics

# File Description

04-deployment — Deployment and Operations

# File Description

Quick Reference

Agent Pattern Selection:
  Simple task          → Single Agent + Tool Use
  Complex task         → Plan-and-Execute
  Mixed specialties    → Multi-Agent
  Human approval needed → Human-in-the-Loop
  Many external tools  → MCP Integration

Framework Selection:
  Lightweight, flexible   → Anthropic SDK directly
  Full-stack              → LangChain / LangGraph
  Multi-agent             → CrewAI / AutoGen
  TypeScript              → Vercel AI SDK

References

  1. Anthropic. "Tool Use Documentation." docs.anthropic.com, 2024.
  2. LangChain. "Documentation." langchain.com/docs, 2024.
  3. Anthropic. "MCP Specification." modelcontextprotocol.io, 2024.

Expand your agent's capabilities with these related and highly-rated skills.

Didn't find tool you were looking for?

Be as detailed as possible for better results