137f1a32fa
Phase 4a: Agent Swarm Collaboration Framework (agent-swarm.ts) - AgentSwarm class with configurable coordinator + specialist agents - Three collaboration modes: Sequential (chain), Parallel (concurrent), Debate (multi-round) - Auto task decomposition based on specialist capabilities - Debate consensus detection with keyword similarity heuristic - Rule-based result aggregation with structured markdown output - Specialist management (add/update/remove) and config updates - History persistence to localStorage (last 25 tasks) - Memory integration: saves task completion as lesson memories Phase 4b: Skill Discovery Engine (skill-discovery.ts) - SkillDiscoveryEngine with 12 built-in skill definitions from skills/ directory - Multi-signal search: name, description, triggers, capabilities, category matching - Conversation-based skill recommendation via topic extraction (CN + EN patterns) - Memory-augmented confidence scoring for suggestions - Skill registration, install status toggle, category filtering - localStorage persistence for skill index and suggestion cache Phase 4c: chatStore Integration - dispatchSwarmTask(description, style): creates and executes swarm task, adds result as message - searchSkills(query): exposes skill search to UI layer Tests: 317 passing across 13 test files (43 new for swarm + skills) - AgentSwarm: createTask, sequential/parallel/debate execution, history, specialist mgmt - SkillDiscovery: search, suggest, register, persist, categories Refs: ZCLAW_AGENT_INTELLIGENCE_EVOLUTION.md updated - all 4 phases complete
ZCLAW 🦞 — OpenClaw 定制版 (Tauri Desktop)
像 AutoClaw (智谱) 和 QClaw (腾讯) 一样，对 OpenClaw 进行定制化封装，打造中文优先的 Tauri 桌面 AI 助手。
核心定位
OpenClaw Gateway (执行引擎)
↕ WebSocket
ZCLAW Tauri App (桌面 UI)
+ 中文模型 Provider (GLM/Qwen/Kimi/MiniMax)
+ 飞书 Channel Plugin
+ 分身(Clone) 管理
+ 自定义 Skills
功能特色
- 基于 OpenClaw: 真实工具执行 (bash/file/browser)、Skills 生态、MCP 协议、心跳引擎
- 中文模型: 智谱 GLM-5、通义千问、Kimi K2.5、MiniMax (OpenAI 兼容 API)
- 飞书集成: 飞书 Channel Plugin，在飞书中直接对话指挥电脑
- 分身系统: 多个独立 Agent 实例，各有自己的角色、记忆、配置
- Tauri 桌面: Rust + React 19，体积小 (~10MB)，性能好
- 设置页面: 对标 AutoClaw — 通用/模型/MCP/技能/IM/工作区/隐私
技术栈
| 层级 | 技术 |
|---|---|
| 执行引擎 | OpenClaw Gateway (Node.js, ws://127.0.0.1:18789) |
| 桌面壳 | Tauri 2.0 (Rust + React 19) |
| 前端 | React 19 + TailwindCSS + Zustand + Lucide Icons |
| 自定义插件 | TypeScript (OpenClaw Plugin API) |
| 通信协议 | OpenClaw Gateway WebSocket Protocol v3 |
项目结构
ZClaw/
├── desktop/ # Tauri 桌面应用 (React 前端)
│ ├── src/
│ │ ├── components/ # UI 组件
│ │ ├── store/ # Zustand 状态管理
│ │ └── lib/gateway-client.ts # Gateway WebSocket 客户端
│ └── src-tauri/ # Rust 后端 (TODO)
│
├── src/gateway/ # Gateway 管理层
│ ├── manager.ts # OpenClaw 子进程管理
│ ├── ws-client.ts # Node.js WebSocket 客户端
│ └── index.ts
│
├── plugins/ # ZCLAW 自定义 OpenClaw 插件
│ ├── zclaw-chinese-models/ # 中文模型 Provider (GLM/Qwen/Kimi/MiniMax)
│ ├── zclaw-feishu/ # 飞书 Channel Plugin
│ └── zclaw-ui/ # UI 扩展 RPC 方法
│
├── skills/ # 自定义 Skills
│ ├── chinese-writing/ # 中文写作
│ └── feishu-docs/ # 飞书文档操作
│
├── config/ # OpenClaw 默认配置
│ ├── openclaw.default.json # Gateway 配置模板
│ ├── SOUL.md # Agent 人格
│ ├── AGENTS.md # Agent 指令
│ ├── IDENTITY.md # Agent 身份
│ └── USER.md # 用户偏好
│
├── scripts/setup.ts # 首次设置脚本
├── docs/ # 文档
│ ├── architecture-v2.md # 架构设计
│ ├── deviation-analysis.md # 偏离分析报告
│ └── autoclaw界面/ # AutoClaw 参考截图
└── src/core/ # [归档] v1 旧代码
快速开始
1. 安装 OpenClaw
# Windows
iwr -useb https://openclaw.ai/install.ps1 | iex
# macOS / Linux
curl -fsSL https://openclaw.ai/install.sh | bash
2. 安装 ZCLAW
git clone https://github.com/xxx/ZClaw.git
cd ZClaw
pnpm install
pnpm setup # 注册插件 + 复制配置
3. 配置 API Key
openclaw configure # 交互式配置
# 或手动编辑 ~/.openclaw/openclaw.json
4. 启动
openclaw gateway # 启动 OpenClaw Gateway
cd desktop && pnpm tauri dev # 启动 Tauri 桌面应用
对标参考
| 产品 | 基于 | IM 渠道 | 桌面框架 |
|---|---|---|---|
| QClaw (腾讯) | OpenClaw | 微信 + QQ | Electron |
| AutoClaw (智谱) | OpenClaw | 飞书 | 自研 |
| ZCLAW (本项目) | OpenClaw | 飞书 (+ 微信/QQ 计划中) | Tauri 2.0 |
文档
License
MIT