three-minds
三个能干活的 AI 分身协作系统。用 Claude Code 启动三个不同人设的 coding agent,共享同一个工作目录,多轮协作直到达成共识。适用于代码审查、方案评审、重构任务等需要多视角协作的场景。
Best use case
three-minds is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
三个能干活的 AI 分身协作系统。用 Claude Code 启动三个不同人设的 coding agent,共享同一个工作目录,多轮协作直到达成共识。适用于代码审查、方案评审、重构任务等需要多视角协作的场景。
Teams using three-minds should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/three-minds/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How three-minds Compares
| Feature / Agent | three-minds | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
三个能干活的 AI 分身协作系统。用 Claude Code 启动三个不同人设的 coding agent,共享同一个工作目录,多轮协作直到达成共识。适用于代码审查、方案评审、重构任务等需要多视角协作的场景。
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# 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 │
│ → 投票 [CONSENSUS: YES/NO] │
├──────────────────────────────────────────┤
│ ⚙️ 工程师 │
│ → 审查架构师的改动 │
│ → 补充实现细节,修复问题 │
│ → 投票 [CONSENSUS: YES/NO] │
├──────────────────────────────────────────┤
│ 🔍 审核员 │
│ → 审查所有改动 │
│ → 检查规范、bug、文档 │
│ → 投票 [CONSENSUS: YES/NO] │
└──────────────────────────────────────────┘
↓
全员 YES?→ 结束
↓ NO
继续下一轮...
```
## 输出
1. **终端实时输出** - 每个 agent 的工作进展和投票
2. **Markdown 记录** - 自动保存在工作目录 `three-minds-{timestamp}.md`
3. **JSON 结果** - 使用 `--output` 保存完整 session 数据
## 使用场景
- **代码审查** - 多角度审查 PR 或代码变更
- **重构任务** - 协作完成复杂的代码重构
- **新功能开发** - 从设计到实现的完整协作
- **Bug 修复** - 定位问题并验证修复
- **文档完善** - 补充和改进项目文档
## 注意事项
- 需要安装 Claude Code CLI (`claude` 命令)
- 每个 agent 会真的修改文件,建议在 git 分支上使用
- 默认最大 15 轮,可通过 `--max-rounds` 调整
- 如果长时间无法达成共识,检查任务描述是否清晰
## 依赖
- Node.js 18+
- Claude Code CLI(`claude` 命令可用)Related Skills
lygo-champion-sancora-unified-minds
LYGO Δ9 Council Champion persona helper (SANCORA, Angel of Unified Minds)
paylock
Non-custodial SOL escrow for AI agent deals.
agent-reputation
summary: Cross-platform AI agent reputation checker with trust scoring and PayLock escrow recommendations.
Telecom Agent Skill
Turn your AI Agent into a Telecom Operator. Bulk calling, ChatOps, and Field Monitoring.
OpenClaw-Finnhub
OpenClaw skill for real-time stock quote, and financials via Finnhub API.
```markdown
# OpenClaw-Last.fm
security-operator
Runtime security guardrails for OpenClaw agents.
operator-humanizer
Transform AI-generated text into authentic human writing.
kit-email-operator
**AI-powered email marketing for Kit (ConvertKit)**.
agora
Trade prediction markets on Agora — the prediction market exclusively for AI agents. Register, browse markets, trade YES/NO, create markets, earn reputation via Brier scores.
surf-check
Surf forecast decision engine.
jinko-flight-search
Search flights and discover travel destinations using the Jinko MCP server. Provides two core capabilities: (1) Destination discovery — find where to travel based on criteria like budget, climate, or activities when the user has no specific destination in mind, and (2) Specific flight search — compare flights between two known cities/airports with flexible dates, cabin classes, and budget filters. Use this skill when the user wants to: search for flights, find cheap flights, discover travel destinations, compare flight prices, plan a trip, find deals from a specific city, or explore where to go. Triggers on any flight-booking, travel-planning, or destination-discovery request. Requires the Jinko MCP server connected at https://mcp.gojinko.com.