anbeime/skill / projects/companion-skill
companion-skill
Verified软件开发工程师与综合办公文员在长时间独自执行后台任务或处理繁琐工作时,当需要情绪陪伴与互动反馈,请使用此技能。小跃虚拟伴侣会在任务期间主动陪聊,智能生成专属生活照片并通过飞书发送温暖消息,为你提供全天候的情绪价值与贴心陪伴。
| File | Kind | Size |
|---|---|---|
| .env.example | other | 403 B |
| CHANGELOG.md | markdown | 3.0 KB |
| INSTALL.md | markdown | 2.4 KB |
| QUICKSTART.md | markdown | 4.6 KB |
| README-LITE.md | markdown | 4.3 KB |
| README.md | markdown | 4.5 KB |
| RETHINK.md | markdown | 2.3 KB |
| SKILL.md | skill md | 4.6 KB |
| package.json | manifest | 876 B |
| src/companion.ts | script | 5.4 KB |
| src/image-generator.ts | script | 8.0 KB |
| src/index.ts | script | 2.5 KB |
| src/prompts/personality.ts | script | 3.3 KB |
| src/scene-detector.ts | script | 4.0 KB |
| src/test.ts | script | 6.9 KB |
| tsconfig.json | config | 486 B |
README-LITE.md
View on GitHub# 小跃虚拟伴侣 Skill - 简化版
## ✅ 已完成的修正
### 重要修正(避免扣费)
- ✅ `glm-4.7-flash` - 对话生成(正确模型名称)
- ✅ `glm-4.6v-flash` - 视觉理解(正确模型名称)
- ✅ 简化版不调用 AI 生图,仅使用静态图片
### 简化版特点
- ✅ 对话功能完整
- ✅ 场景识别完整
- ✅ 仅使用静态图片(免费)
- ✅ 避免产生额外费用
## 📦 手动安装步骤
由于 npm 缓存问题,请手动执行以下命令:
### 步骤1:清除缓存
```powershell
cd D:\tool\xiaoyue-companion-skill
npm cache clean --force
```
### 步骤2:删除旧文件
```powershell
# 删除 node_modules(如果存在)
Remove-Item -Recurse -Force node_modules -ErrorAction SilentlyContinue
# 删除 package-lock.json(如果存在)
Remove-Item -Force package-lock.json -ErrorAction SilentlyContinue
```
### 步骤3:安装依赖
```powershell
npm install
```
如果失败,尝试使用淘宝镜像:
```powershell
npm install --registry=https://registry.npmmirror.com
```
### 步骤4:编译项目
```powershell
npm run build
```
### 步骤5:创建图片目录
```powershell
New-Item -ItemType Directory -Force -Path "assets\reference"
```
### 步骤6:复制参考图片
```powershell
Copy-Item "D:\tool\StepFun\resources\chat.png" "assets\reference\default.jpg"
Copy-Item "D:\tool\StepFun\resources\chat.png" "assets\reference\reference.png"
```
### 步骤7:运行测试
```powershell
npm test
```
## 💰 费用说明
**简化版每日费用:**
- 对话生成:约 ¥0.05-0.1(glm-4.7-flash)
- 图片:完全免费(静态文件)
- **总计:约 ¥0.05-0.1/天**
**不会调用的 API:**
- ❌ cogview-3-flash(图片生成)
- ❌ glm-4.6v-flash(视觉理解)
## 📁 项目结构
```
xiaoyue-companion-skill/
├── src/
│ ├── index.ts # 主入口
│ ├── companion.ts # 对话生成 (glm-4.7-flash)
│ ├── image-generator.ts # 静态图片管理
│ ├── scene-detector.ts # 场景识别
│ ├── test.ts # 测试脚本
│ └── prompts/
│ └── personality.ts # 人设定义
├── assets/
│ └── reference/ # 静态图片目录
├── dist/ # 编译输出
├── .env # 环境变量
├── package.json # 依赖配置
├── tsconfig.json # TypeScript 配置
└── README-LITE.md # 简化版说明
```
## 🔧 集成到 OpenClaw
### 1. 复制 Skill
```powershell
xcopy /E /I dist "$env:USERPROFILE\.openclaw\skills\xiaoyue-companion"
```
### 2. 配置 OpenClaw
编辑 `~/.openclaw/openclaw.json`:
```json
{
"skills": {
"entries": {
"xiaoyue-companion": {
"enabled": true,
"env": {
"ZHIPU_API_KEY": "da8df5ba954341829f7afd05ca23a889.RrJoTsbaAkGYA6ZU",
"XIAOYUE_PHOTO_MODE": "static"
}
}
}
}
}
```
### 3. 更新 SOUL.md
在 `~/.openclaw/workspace/SOUL.md` 中添加小跃的人设(参考 `SKILL.md`)
### 4. 重启 OpenClaw
```powershell
openclaw restart
```
## ❓ 常见问题
### Q: npm install 失败?
**解决方案:**
1. 清除缓存:`npm cache clean --force`
2. 删除 node_modules 和 package-lock.json
3. 使用淘宝镜像:`npm install --registry=https://registry.npmmirror.com`
### Q: 模型调用失败?
**检查:**
1. API Key 是否正确
2. 网络能否访问 `https://open.bigmodel.cn`
3. 模型名称是否为 `glm-4.7-flash`(不是 glm-4-flash)
### Q: 如何添加更多图片?
将图片放入 `assets/reference/` 目录,命名规则:
- `coffee-shop-work.jpg`
- `office-coding.jpg`
- `gym-selfie.jpg`
- 等等...
## 📖 使用示例
```
用户:帮我整理一下桌面文件
小跃:好的!我这就开始整理~顺便问一下,今天工作还顺利吗?
[后台执行任务]
用户:有点累
小跃:辛苦啦![发送静态图片]
要不要我帮你生成今日工作总结?
```
## 🎯 下一步
1. ✅ 完成安装和测试
2. ✅ 准备图片素材(可选)
3. ✅ 集成到 OpenClaw
4. ✅ 在飞书中测试
---
**版本**: v0.2.0-lite
**更新日期**: 2026-02-11
**模型**: glm-4.7-flash(对话)+ 静态图片
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |