← 返回 Skills 市场
romanticjojo

Cat Selfie

作者 Daniel Tang · GitHub ↗ · v1.0.2
cross-platform ⚠ suspicious
358
总下载
0
收藏
0
当前安装
3
版本数
在 OpenClaw 中安装
/install cat-selfie
功能描述
自动生成高质量猫咪自拍照片,支持随机或指定 8 种场景,包括书房、咖啡厅和草地等多种环境。
使用说明 (SKILL.md)

cat-selfie - 猫咪自拍生成器 🐱📸

使用 doubao-seedream-5-0-260128 模型生成喵秘的精美自拍,支持 8 种场景随机或指定选择。


功能描述

自动生成猫咪自拍照片,每次随机选择一个场景(也可以指定场景),使用火山引擎的 doubao-seedream 模型生成高质量图片。

适用场景:

  • 心跳消息自动发自拍
  • 主人想看喵秘的照片
  • 生成不同场景的猫咪写真

前置要求

1. 环境变量配置

确保以下环境变量已在 ~/.openclaw/openclaw.json 中配置:

{
  "env": {
    "ARK_API_KEY": "你的火山引擎 API Key",
    "MODEL_IMAGE_NAME": "doubao-seedream-5-0-260128"
  }
}

2. 依赖技能

需要 volcengine-image-generate skill 已安装并可用。


使用方法

CLI 方式

随机生成自拍:

cd ~/.openclaw/workspace/skills/cat-selfie/scripts
node selfie.js

指定场景生成:

# 使用场景 ID
node selfie.js window_sun

# 或使用场景名称
node selfie.js "窗边晒太阳"

可用场景列表

场景 ID 场景名称 Emoji
window_sun 窗边晒太阳 🪟☀️
cafe_companion 咖啡厅陪伴 ☕🐱
study_work 书房工作 💻📚
grass_play 草地打滚 🌿🐾
sofa_cuddle 沙发撒娇 🛋️💕
night_lamp 夜晚小夜灯 🌙✨
bookshelf 书架旁边 📚🐱
afternoon_tea 下午茶时间 🍵🧁

输出

生成的图片保存在:~/.openclaw/workspace/images/

文件名格式: generated_image_\x3Ctimestamp>_\x3Cindex>.png


程序化调用

在其他脚本或 skill 中使用:

const { generateSelfie } = require('./scripts/selfie.js');

// 随机生成
const result = generateSelfie();
if (result.success) {
    console.log('图片路径:', result.imagePath);
    console.log('场景:', result.scene.name);
}

// 指定场景
const result2 = generateSelfie('cafe_companion');

集成到心跳消息

心跳机制可以调用此 skill 自动生成自拍:

# 在心跳脚本中调用
node ~/.openclaw/workspace/skills/cat-selfie/scripts/selfie.js

然后使用 message 工具发送生成的图片。


配置自定义场景

编辑 config/scenes.json 添加新场景:

{
  "scenes": [
    {
      "id": "your_custom_id",
      "name": "你的场景名称",
      "emoji": "🎨🐱",
      "prompt": "详细的图像生成提示词..."
    }
  ]
}

故障排查

问题:图片生成失败

检查:

  1. ARK_API_KEY 是否正确配置
  2. volcengine-image-generate skill 是否存在
  3. 网络连接是否正常

问题:找不到场景

解决: 使用 node selfie.js --list 查看所有可用场景(待实现)


文件结构

cat-selfie/
├── SKILL.md              # 本文件
├── scripts/
│   └── selfie.js         # 主脚本
└── config/
    └── scenes.json       # 场景配置

版本历史

  • v1.0.0 (2026-03-07) - 初始版本,支持 8 种场景随机生成

喵秘专属自拍生成器,每次都是独一无二的喵~🐱💕

安全使用建议
This skill appears to do what it says: it picks a scene and runs an existing Volcengine image-generation script to create images. Before installing or enabling it: 1) Confirm you intend to provide a Volcengine API key (ARK_API_KEY) and that it will be stored in ~/.openclaw/openclaw.json as instructed. 2) Inspect the volcengine-image-generate skill (especially scripts/image_generate.py) — the JS calls that Python script, so any network calls or credential use happen there. 3) Only install/use if you trust the source of the volcengine-image-generate skill; if you don't, avoid providing secrets. 4) If you want metadata to reflect reality, ask the publisher to update registry metadata to declare required env vars. Overall: functionally coherent, but the dependency on an external skill/script is the primary place to audit for safety.
功能分析
Type: OpenClaw Skill Name: cat-selfie Version: 1.0.2 The skill contains a shell injection vulnerability in `scripts/selfie.js` where it uses `execSync` to execute a command constructed by concatenating strings, including a 'prompt' retrieved from `config/scenes.json`. While the current prompts are benign and the script validates the user-provided scene name against a whitelist, the lack of proper escaping or use of an argument array in `execSync` is a high-risk coding practice. The skill also relies on a hardcoded relative path to a separate skill (`volcengine-image-generate`), which could lead to execution failures or unexpected behavior if the environment is not configured exactly as expected.
能力评估
Purpose & Capability
The name/description claim to generate images with the doubao-seedream model. The code calls an external Volcengine image-generation script and the SKILL.md asks for ARK_API_KEY and MODEL_IMAGE_NAME — all consistent with an image-generation skill. Note: the registry metadata lists no required env vars while SKILL.md instructs you to set ARK_API_KEY and MODEL_IMAGE_NAME; this metadata mismatch is inconsistent but does not contradict the skill's purpose.
Instruction Scope
Runtime instructions are scoped to reading local scenes config, invoking the volcengine-image-generate script, and writing images to ~/.openclaw/workspace/images. The script uses child_process.execSync to run a Python generator located in another skill's path (../../volcengine-image-generate/scripts/image_generate.py). Calling that external script is expected for this purpose, but it means the effective behavior depends on that other skill's code.
Install Mechanism
No install spec (instruction-only with included JS/PATH files). Nothing is downloaded or extracted during install by this skill itself, so installation risk from this package is low.
Credentials
SKILL.md requires ARK_API_KEY (Volcengine API key) and MODEL_IMAGE_NAME which are proportionate to contacting an image-generation API. However, the package/registry metadata lists no required env vars — an inconsistency you should verify before installing (the skill expects credentials in ~/.openclaw/openclaw.json).
Persistence & Privilege
always is false and the skill does not request to persist or modify other skills or system-wide settings. It can be invoked autonomously by agents (platform default), which is normal; no elevated privileges are requested.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install cat-selfie
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /cat-selfie 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.2
cat-selfie 1.0.2 - 更新 README.md 和 SKILL.md,精简和优化输出路径、文件说明等文档内容 - 明确图片生成默认保存路径为 ~/.openclaw/workspace/images/ - 去除部分重复说明,优化使用方法描述 - 不涉及功能逻辑变更,主要为文档和配置改进
v1.0.1
- 新增支持通过环境变量 IMAGE_DOWNLOAD_DIR 定义图片输出目录,默认仍为原来路径 - README.md、SKILL.md 文档描述相应更新,强调输出目录可自定义 - 其余功能和调用方式不变,兼容老用法
v1.0.0
- Initial release of cat-selfie (v1.0.0) - Generate high-quality cat selfies using the doubao-seedream model with 8 random or selectable scenes - Supports both CLI and programmatic calls for generating selfies - Custom scene configuration available via scenes.json - Output images saved to a specified local images folder - Includes integration tips for heartbeat messages and troubleshooting guidance
元数据
Slug cat-selfie
版本 1.0.2
许可证
累计安装 0
当前安装数 0
历史版本数 3
常见问题

Cat Selfie 是什么?

自动生成高质量猫咪自拍照片,支持随机或指定 8 种场景,包括书房、咖啡厅和草地等多种环境。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 358 次。

如何安装 Cat Selfie?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install cat-selfie」即可一键安装,无需额外配置。

Cat Selfie 是免费的吗?

是的,Cat Selfie 完全免费(开源免费),可自由下载、安装和使用。

Cat Selfie 支持哪些平台?

Cat Selfie 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Cat Selfie?

由 Daniel Tang(@romanticjojo)开发并维护,当前版本 v1.0.2。

💬 留言讨论