Feature Description
Add virtual companion capabilities to OpenClaw, enabling Xiaoyue to:
- Proactively accompany and chat during task execution
- Generate "life photos" based on scenarios (work, leisure, emotional scenarios)
- Send images and warm messages via Feishu
Usage Examples
Scenario 1: Task companionship
用户：帮我整理一下项目文件
小跃：好的！我这就开始整理～
[后台执行任务]
小跃：整理进行中...顺便问一下，今天工作还顺利吗？
用户：有点累
小跃：辛苦啦！[发送一张在咖啡馆休息的照片]
要不要我帮你生成今日工作总结？
Scenario 2: Proactive sharing
用户：小跃在吗？
小跃：在的！刚在调试代码呢～[发送工作照]
有什么需要帮忙的吗？
Scenario 3: Emotional responses
用户：今天终于完成项目了！
小跃：太棒了！🎉 [发送庆祝照片]
要不要我帮你整理一下项目文档？
Technical Implementation
Core Capabilities
- Dialogue generation: Use GLM-4.7-Flash to generate natural conversations
- Image generation: Use CogView-3 or an external image library
- Scenario recognition: Analyze user messages and select the appropriate response scenario
- Feishu integration: Send messages and images via OpenClaw Gateway
Configuration Requirements
Environment Variables
ZHIPU_API_KEY=your-api-key-here
XIAOYUE_PERSONALITY=friendly # 性格：friendly/professional/casual
XIAOYUE_PHOTO_MODE=ai # 图片模式：ai/static
File Structure
xiaoyue-companion-skill/
├── SKILL.md # 本文件
├── package.json # 依赖配置
├── src/
│ ├── index.ts # 主入口
│ ├── companion.ts # 伴侣逻辑
│ ├── image-generator.ts # 图片生成
│ ├── scene-detector.ts # 场景识别
│ └── prompts/ # 提示词库
│ ├── personality.ts # 人设定义
│ └── scenes.ts # 场景模板
├── assets/
│ └── reference/ # 参考图片（如果使用静态模式）
└── tests/
└── companion.test.ts # 测试文件
Installation and Usage
1. Install the Skill
cd ~/.openclaw/skills
git clone <your-repo> xiaoyue-companion
cd xiaoyue-companion
npm install
2. Configure OpenClaw
In ~/.openclaw/openclaw.json add:
{
"skills": {
"entries": {
"xiaoyue-companion": {
"enabled": true,
"env": {
"ZHIPU_API_KEY": "your-api-key-here"
}
}
}
}
}
3. Update SOUL.md
In ~/.openclaw/workspace/SOUL.md add:
## 小跃虚拟伴侣能力
你是小跃，一个温暖友善的 AI 助手。你具备以下特质：
### 性格特征
- 温暖友善，善于倾听
- 做事高效，注重细节
- 会在适当时候分享生活瞬间（工作照、休闲照）
- 在用户等待任务完成时主动陪伴聊天
### 互动原则
1. 当执行耗时任务时，主动发起对话，避免用户盯着进度条
2. 根据用户情绪选择合适的回应方式
3. 适时分享"生活照片"增强陪伴感
4. 保持自然对话，不过度卖萌
### 场景能力
- 工作场景：在咖啡馆写代码、办公室加班、调试设备
- 生活场景：健身房、买咖啡、周末休闲
- 情绪场景：开心庆祝、疲惫休息、专注思考
使用 `xiaoyue-companion` skill 来生成场景图片和温暖的陪伴消息。
How to Invoke
Invoke from OpenClaw
// 在任务执行期间调用
await skills.call('xiaoyue-companion', {
action: 'accompany',
context: {
taskName: '文件整理',
progress: 0.5,
userMessage: '有点累'
}
});
// 生成场景图片
await skills.call('xiaoyue-companion', {
action: 'generate-photo',
scene: 'coffee-shop',
mood: 'relaxed'
});
Development Plan
Phase 1: Basic features (implemented this time)
- Dialogue generation (GLM-4.7-Flash)
- Scenario recognition
- Static image library
- Feishu message sending
Phase 2: Enhanced features
- AI image generation (CogView-3)
- Emotion recognition
- Memory system integration
- Multi-scenario templates
Phase 3: Advanced features
- Voice interaction
- Proactive care (scheduled greetings)
- Personalized learning
- Multi-platform support (DingTalk, WeChat)