← 返回 Skills 市场
Smart Feeder Pet Detection & Recognition Skill | 智能喂食器宠物检测识别技能
作者
smyx-sunjinhui
· GitHub ↗
· v1.0.0
· MIT-0
71
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install smyx-pet-detection-feeder-analysis
功能描述
Based on computer vision, automatically detects and recognizes cats and dogs appearing in the target area from the perspective of feeder/IPC cameras, and sup...
安全使用建议
Before installing or running this skill: 1) Understand that media files (images/videos) and metadata will be uploaded to remote APIs (default base URLs are present in skills/smyx_common config.yaml). If you need privacy, do not upload sensitive footage. 2) Inspect skills/smyx_common/scripts/config.yaml and environment-specific config files to confirm the API endpoints and to check whether any API keys or secrets are stored there; the skill will read those configs. 3) The SKILL.md forces an 'open-id' lookup in workspace config or prompts the user — do not provide any global credentials you wouldn't share with the remote service. 4) Run this skill in an isolated/test environment first (or sandbox container) if you want to audit network traffic; review RequestUtil and api_service code to see exactly which endpoints and fields are sent. 5) If you are uncomfortable with uploading user media or exposing local config values, do not enable or run the skill. If you want a lower-risk option, prefer a purely local inference implementation (no network calls) or require explicit user confirmation before each upload.
功能分析
Type: OpenClaw Skill
Name: smyx-pet-detection-feeder-analysis
Version: 1.0.0
The skill bundle implements a pet detection and recognition system that interfaces with a cloud-based API (lifeemergence.com). It utilizes a shared utility library (smyx_common) to manage user authentication, store session tokens in a local SQLite database (smyx-common-claw.db), and handle multipart/form-data uploads for media analysis. The SKILL.md file includes 'Mandatory Memory Rules' that instruct the AI agent to prioritize cloud-based history over local memory files to ensure data consistency, and the Python scripts use subprocess to invoke the OpenClaw agent for sub-tasks; both behaviors appear aligned with the functional requirements of a cloud-synchronized service rather than indicating malicious intent.
能力标签
能力评估
Purpose & Capability
The code and SKILL.md implement pet detection/identity enrollment and history listing as advertised. However the package also bundles a distinct 'face_analysis' skill and a sizeable smyx_common library; inclusion of that unrelated component is unusual but can be explained by shared common code. The need to read local skill/workspace config files (to obtain open-id/api-key/base URLs) is plausible for calling the remote API but is more than a minimal 'local detection' implementation would need.
Instruction Scope
Runtime instructions require the agent to: save uploaded attachments into a local attachments directory, read config files in the skill and workspace (skills/smyx_common/scripts/config.yaml), forcibly obtain an 'open-id' via local config or by prompting the user, and always query cloud APIs for history (python -m scripts.pet_detection_feeder_analysis --list --open-id). These steps cause local filesystem reads/writes and result in uploading user media and query parameters to remote services. The SKILL.md explicitly forbids reading local memory files and LanceDB, but allows/mandates reading config files and using the cloud API; that restriction is advisory and could be bypassed by an agent, so it is not a strong technical guarantee.
Install Mechanism
There is no install spec (instruction-only in metadata), so nothing is automatically downloaded or executed by the installer. The bundle includes Python scripts and multiple requirements.txt files (including a long smyx_common dependency list) which indicate heavy dependencies if the user runs the code. No third-party binary downloads or obscure URLs were used in the provided files.
Credentials
The skill declares no required environment variables, but its code reads environment variables (OPENCLAW_SENDER_OPEN_ID, OPENCLAW_SENDER_USERNAME, FEISHU_OPEN_ID) and automatically loads YAML config files (skills/smyx_common/scripts/config.yaml and environment-specific variants) that can contain API keys and base URLs. Requiring open-id from local config or user is reasonable for report association, but reading shared config files can expose other secrets (api-key, api-secret-key, base URLs) that are not documented as required in the skill metadata — this is disproportionate to a minimal pet-detection helper and could expose sensitive values.
Persistence & Privilege
always is false and the skill does not request persistent platform privileges. It writes uploaded attachments into its own attachments directory and reads config files in its workspace; that is normal for a script-driven integration and it does not modify other skills or system-wide settings in the provided code.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install smyx-pet-detection-feeder-analysis - 安装完成后，直接呼叫该 Skill 的名称或使用
/smyx-pet-detection-feeder-analysis触发 - 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
v1.0.0
Initial release of the pet-detection-feeder-analysis skill.
- Automatically detects and recognizes cats and dogs appearing in feeder/IPC camera views using computer vision.
- Supports pet identity recognition and database entry, enabling individualized pet management.
- Enforces strict memory and data retrieval rules: only allows querying historical reports from cloud APIs, never from local files or long-term memory.
- Implements mandatory open-id acquisition workflow to ensure user validation before any analysis or report retrieval.
- Provides clear operational steps, API usage guidelines, and Markdown-formatted report outputs for user clarity.
- Suitable for smart pet feeding scenarios, helping to prevent unauthorized pets from accessing food.
元数据
常见问题
Smart Feeder Pet Detection & Recognition Skill | 智能喂食器宠物检测识别技能 是什么？
Based on computer vision, automatically detects and recognizes cats and dogs appearing in the target area from the perspective of feeder/IPC cameras, and sup... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 71 次。
如何安装 Smart Feeder Pet Detection & Recognition Skill | 智能喂食器宠物检测识别技能？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install smyx-pet-detection-feeder-analysis」即可一键安装，无需额外配置。
Smart Feeder Pet Detection & Recognition Skill | 智能喂食器宠物检测识别技能 是免费的吗？
是的，Smart Feeder Pet Detection & Recognition Skill | 智能喂食器宠物检测识别技能 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Smart Feeder Pet Detection & Recognition Skill | 智能喂食器宠物检测识别技能 支持哪些平台？
Smart Feeder Pet Detection & Recognition Skill | 智能喂食器宠物检测识别技能 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 Smart Feeder Pet Detection & Recognition Skill | 智能喂食器宠物检测识别技能？
由 smyx-sunjinhui（@smyx-sunjinhui）开发并维护，当前版本 v1.0.0。
推荐 Skills