three-minds

# Three Minds - 三个臭皮匠顶个诸葛亮

真正能干活的多 Agent 协作系统。不是纸上谈兵，而是真的能读文件、改代码、跑测试。

## 核心特性

- 🔧 **真正执行** - 每个 agent 都能读取文件、编写代码、运行测试
- 👥 **多视角协作** - 三个不同人设的 agent 互相审核
- ✅ **共识机制** - 全员 YES 才结束，确保质量
- 📁 **共享目录** - 在同一个项目上协作
- 📝 **完整记录** - 自动保存讨论和改动历史

## 使用方法

```bash
# 基础用法
three-minds "审查并改进这个项目的代码质量" --dir ./my-project

# 使用代码审查配置（安全+性能+质量三人组）
three-minds "审查 src/ 目录的所有代码" --config code-review --dir ./project

# 指定最大轮数
three-minds "重构这个模块" --dir ./module --max-rounds 5

# 保存结果到 JSON
three-minds "任务描述" --dir ./project --output result.json
```

## 触发词

- "三个臭皮匠"
- "three minds"
- "多 agent 协作"
- "让三个 agent 一起审查"
- "协作完成这个任务"

## 预设配置

### 默认配置 - 代码协作三人组
- 🏗️ **架构师** - 关注代码结构、设计模式、可扩展性
- ⚙️ **工程师** - 关注代码质量、错误处理、性能优化
- 🔍 **审核员** - 关注代码规范、潜在 bug、文档完整性

### code-review - 代码审查三人组
- 🛡️ **安全专家** - 关注安全漏洞、注入风险、权限问题
- ⚡ **性能工程师** - 关注算法复杂度、内存使用、查询优化
- ✅ **质量审查员** - 关注可读性、命名规范、测试覆盖

## 自定义配置

创建 JSON 配置文件：

```json
{
  "name": "我的三人组",
  "agents": [
    {
      "name": "专家A",
      "emoji": "🎯",
      "persona": "你是一位...关注..."
    },
    {
      "name": "专家B",
      "emoji": "🔬",
      "persona": "你是一位...擅长..."
    },
    {
      "name": "专家C",
      "emoji": "📊",
      "persona": "你是一位...负责..."
    }
  ],
  "maxRounds": 10,
  "projectDir": "."
}
```

然后：`three-minds "任务" --config ./my-config.json`

## 工作流程

```
┌──────────────────────────────────────────┐
│              第 N 轮                      │
├──────────────────────────────────────────┤
│  🏗️ 架构师                               │
│  → 读取文件，审查结构                     │
│  → 执行必要的重构...

# Three Minds 🧠🧠🧠

**三个臭皮匠顶个诸葛亮** - A Multi-Agent Collaboration System

Three AI agents with different personas working together on the same codebase. Not just talking—they actually read files, write code, and run tests.

## Features

- 🔧 **Real Execution** - Each agent can read files, write code, run tests via Claude Code CLI
- 👥 **Multi-Perspective** - Three agents with different expertise review each other's work
- ✅ **Consensus Voting** - All must vote YES to finish, ensuring quality
- 📁 **Shared Workspace** - Collaborate on the same project directory
- 📝 **Full Transcript** - Auto-saves discussion and changes history

## Installation

```bash
# Clone
git clone https://github.com/Enderfga/three-minds.git
cd three-minds

# Install dependencies
npm install

# Build
npm run build

# Link globally (optional)
npm link
```

## Requirements

- Node.js 18+
- **Claude Code CLI** (`claude` command must be available)

## Usage

```bash
# Basic usage
three-minds "Review and improve this project's code quality" --dir ./my-project

# Use code-review preset (security + performance + quality trio)
three-minds "Review all code in src/" --config code-review --dir ./project

# Specify max rounds
three-minds "Refactor this module" --dir ./module --max-rounds 5

# Save result to JSON
three-minds "task description" --dir ./project --output result.json
```

## Preset Configurations

### Default - Code Collaboration Trio
- 🏗️ **Architect** - Code structure, design patterns, scalability
- ⚙️ **Engineer** - Code quality, error handling, performance
- 🔍 **Reviewer** - Code standards, potential bugs, documentation

### code-review - Code Review Trio
- 🛡️ **Security Expert** - Vulnerabilities, injection risks, permissions
- ⚡ **Performance Engineer** - Algorithm complexity, memory, query optimization
- ✅ **Quality Reviewer** - Readability, naming conventions, test coverage

### idea-brainstorm - Research Brainstorm Trio
- 📚 **Literature Expert** - Related work, theoret...