XiangwenXiao

Xxw Agent — Development skill for Claude Code

Development community

基于python,参考claude code制作的一个小型通用agent,如果你想快速学习claude code的做法,但又苦于claude code代码太多,又不熟悉TS,可以fork本仓库用于学习.

How to install Xxw Agent

This entry records only its repository, not the path inside it, so there is no exact command to give. Open XiangwenXiao/xxw_agent and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Xxw Agent does

基于python,参考claude code制作的一个小型通用agent,如果你想快速学习claude code的做法,但又苦于claude code代码太多,又不熟悉TS,可以fork本仓库用于学习.

Alternatives in Development

  • Auto Update — Pull the latest ECC repo changes and reinstall the current managed targets 243.5k ★
  • Sharing Skills — skill for guidance 115.1k ★
  • Template Skill — Minimal skeleton for a new skill project structure 97.5k ★

README

🤖 Minimal Agent

**A lightweight, Claude Code-inspired Agent implementation in Python**

[![Python 3.10+](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/downloads/) [![Anthropic API](https://img.shields.io/badge/Anthropic-API-orange.svg)](https://www.anthropic.com/api) [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)

*Built with streaming responses · Concurrent tool execution · Smart context compression*


🌐 **English** | [中文](README-zh.md)


📋 Table of Contents


✨ Features

Core Capabilities

Feature Description Status
🔄 Agent Loop Automatic LLM → Tool → Result iteration ✅ Ready
⚡ Streaming Real-time token-by-token output ✅ Ready
🛠️ Tools Bash, Read, Write, Web Search, Ask, Todo ✅ Ready
🏃 Concurrency Parallel tool execution with mutex groups ✅ Ready
🌐 Web Search Bing + DuckDuckGo, no API key needed ✅ Ready
📦 Context Compression Three-layer smart compression ✅ Ready
💾 Memory Persistent user/project/reference storage ✅ Ready
📋 Todo Reminder Periodic task list reminders ✅ Ready
📝 Logging Python logging module, file-only output ✅ Ready

Code Quality

  • 🎯 Type Hints: Full type annotation coverage
  • 🧪 Async/Await: Modern async patterns throughout
  • 📐 Modular Design: Clean separation of concerns
  • 🔒 Error Handling: Graceful failure recovery

🚀 Quick Start

Prerequisites

  • Python 3.1