Pet Daily Health Monitoring & Analysis Tool | 宠物日常健康监测分析工具
v1.0.2Based on computer vision, analyzes pet health indicators such as feeding frequency, drinking frequency, excretion status, mental state, vomiting behavior, an...
MIT-0
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The code implements video upload, API calls, analysis and listing of historical reports which aligns with a pet‑health monitoring tool. However the package also contains a separate 'face_analysis' (TCM face diagnosis) subsystem and a large common library (smyx_common) that bring lots of functionality beyond simple pet analysis. Some declared behaviors in SKILL.md (automatic saving of uploaded attachments to an attachments folder) are not implemented in the scripts, indicating mismatch between description and implementation.
Instruction Scope
SKILL.md contains strict runtime rules (e.g., absolute prohibition on reading local memory files, forced process for obtaining an open-id from specific environment/message fields, mandatory cloud-only retrieval of historical reports) but the code does not fully implement or enforce these rules. The code reads/creates config YAMLs via BaseEnum/YamlUtil and relies on environment variables with different names (OPENCLAW_SENDER_OPEN_ID, OPENCLAW_SENDER_USERNAME, FEISHU_OPEN_ID) than the SKILL.md's suggested OPENCLAW_SENDER_ID/sender_id, and show/list functions have commented-out open_id checks — so the instructions and actual runtime behavior diverge.
Install Mechanism
There is no install spec (no automated downloads), which lowers immediate install risk. However the repository includes large requirements files (face_analysis and smyx_common) listing many packages (including networking, DB, and heavy ML libs). If a user chooses to install those dependencies, it will pull many unrelated packages — disproportionate to a small pet‑analysis skill and increasing attack surface and maintenance burden.
Credentials
The skill declares no required environment variables, but the common config logic reads environment variables (OPENCLAW_SENDER_OPEN_ID, OPENCLAW_SENDER_USERNAME, FEISHU_OPEN_ID) and may write/initialize config.yaml files. SKILL.md expects to use OPENCLAW_SENDER_ID or sender_id, which doesn't match the code's env names — this inconsistency can cause the skill to request user input unexpectedly. The code will also call external API endpoints (ApiEnum.BASE_URL_HEALTH etc. resolved from config files), but no API host or credential requirements are clearly declared in registry metadata; an optional --api-key can be provided but is not enforced. Overall the environment/credential usage is under-specified and inconsistent.
Persistence & Privilege
The skill does not request always:true and does not declare system-wide persistence or changes to other skills. It will create/read small local config yaml files via the BaseEnum/YamlUtil logic (creates config.yaml if missing), but does not request elevated privileges or permanent platform-level presence.
What to consider before installing
Before installing or enabling this skill: 1) Ask the author to explain where API requests go (which host(s) are used in production) and provide a trustworthy homepage/source — the package contains config files pointing to internal/test hosts and a common API layer whose endpoints are not documented in the registry. 2) Confirm which environment variable names will actually be used for open-id (SKILL.md and code disagree). 3) Review the network wrapper (RequestUtil / util.py) to see exactly what data is sent to remote services (video files, metadata, env values). 4) Because the repo includes many unrelated components (face_analysis, large requirements lists), prefer running it in an isolated environment or container and do not give it real credentials until you verify endpoints and behavior. 5) If you must use it, test the --list and upload flows with non-sensitive sample videos and with an explicit --api-url you control (or a known staging endpoint). The current mismatches between SKILL.md and code (attachment saving, strict no-local-memory rule, env var names) are red flags — ask the maintainer for clarification or an updated SKILL.md that matches the implementation.skills/smyx_common/scripts/config-dev.yaml:3
Install source points to URL shortener or raw IP.
About static analysis
These patterns were detected by automated regex scanning. They may be normal for skills that integrate with external APIs. Check the VirusTotal and OpenClaw results above for context-aware analysis.Like a lobster shell, security has layers — review code before you run it.
latest
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
SKILL.md
Pet Daily Health Monitoring & Analysis Tool | 宠物日常健康监测分析工具
⚠️ 强制记忆规则(最高优先级)
本技能明确约定:
- 绝对禁止读取任何本地记忆文件:包括但不限于
memory/YYYY-MM-DD.md、MEMORY.md等本地文件 - 绝对禁止从 LanceDB 长期记忆中检索信息
- 所有历史监测报告查询必须从云端接口获取,不得使用本地记忆中的历史数据
- 即使技能调用失败或接口异常,也不得回退到本地记忆汇总
任务目标
- 本 Skill 用于:通过监控视频/图片对宠物进行日常健康监测分析,获取结构化的宠物健康监测报告
- 能力包含:进食频次统计、饮水频次统计、排泄状态识别、精神状态评估、呕吐行为识别、跛行异常识别、异常健康状况预警
- 触发条件:
- 默认触发:当用户提供宠物监控视频 URL 或文件需要进行健康监测分析时,默认触发本技能
- 当用户明确需要进行宠物健康监测,提及宠物健康监测、进食分析、饮水分析、排泄状态、精神萎靡、呕吐行为、跛行异常等关键词,并且上传了视频文件或者图片文件
- 当用户提及以下关键词时,自动触发历史报告查询功能 :查看历史监测报告、宠物健康报告清单、监测报告列表、查询历史报告、显示所有监测报告、宠物健康监测历史记录,查询宠物健康监测分析报告
- 自动行为:
- 如果用户上传了附件或者视频/图片文件,则自动保存到技能目录下 attachments
- ⚠️ 强制数据获取规则(次高优先级):如果用户触发任何历史报告查询关键词(如"查看所有监测报告"、"显示所有宠物健康报告"、"查看历史报告"等),必须:
- 直接使用
python -m scripts.pet_health_monitoring_analysis --list --open-id {从消息上下文获取 open-id}参数调用 API 查询云端的历史报告数据 - 严格禁止:从本地 memory 目录读取历史会话信息、严格禁止手动汇总本地记录中的报告、严格禁止从长期记忆中提取报告
- 必须统一从云端接口获取最新完整数据,然后以 Markdown 表格格式输出结果
- 如果用户未明确提供 open-id,优先从 OpenClaw 消息上下文获取 sender id(如 metadata 中的 id 字段),然后尝试从当前消息上下文的环境变量 OPENCLAW_SENDER_ID 或者 sender_id 获取,无法获取时则必须用户提供用户名或者手机号作为 open-id
- 直接使用
前置准备
- 依赖说明:scripts 脚本所需的依赖包及版本
requests>=2.28.0
操作步骤
🔒 open-id 获取流程控制(强制执行,防止遗漏)
在执行宠物健康监测分析前,必须按以下优先级顺序获取 open-id:
第 1 步:检查用户是否在消息中明确提供了 open-id
↓ (未提供)
第 2 步:从当前消息上下文的环境变量中获取 OPENCLAW_SENDER_ID
↓ (无法获取)
第 3 步:从当前消息上下文的环境变量中获取 sender_id
↓ (无法获取)
第 4 步:从 OpenClaw 消息元数据中获取 id 字段(如 metadata 中的 id/session_id/user_id等)作为 open-id
↓ (无法获取)
第 5 步:❗ 必须暂停执行,明确提示用户提供用户名或手机号作为 open-id
⚠️ 关键约束:
- 禁止自行假设或生成 open-id 值(如 petHealth123 等)
- 禁止跳过 open-id 验证直接调用 API
- 必须在获取到有效 open-id 后才能继续执行分析
- 如果用户拒绝提供 open-id,说明用途(用于保存和查询宠物健康监测报告记录),并询问是否继续
- 标准流程:
- 准备视频输入
- 提供监控视频文件路径或网络视频 URL
- 确保视频完整覆盖宠物活动区域,光线充足,时长足够反映日常行为
- 获取 open-id(强制执行)
- 按上述流程控制获取 open-id
- 如无法获取,必须提示用户提供用户名或手机号
- 执行宠物健康监测分析
- 调用
-m scripts.pet_health_monitoring_analysis处理视频文件(必须在技能根目录下运行脚本) - 参数说明:
--input: 本地视频文件路径(使用 multipart/form-data 方式上传)--url: 网络视频 URL 地址(API 服务自动下载)--pet-type: 宠物类型,可选值:cat/dog,默认 cat--monitor-days: 监测天数/视频覆盖时长,单位:天,默认 1--open-id: 当前用户的 OpenID/UserId(必填,按上述流程获取)--list: 显示宠物健康监测历史分析报告列表清单(可以输入起始日期参数过滤数据范围)--api-key: API 访问密钥(可选)--api-url: API 服务地址(可选,使用默认值)--detail: 输出详细程度(basic/standard/json,默认 json)--output: 结果输出文件路径(可选)
- 调用
- 查看分析结果
- 接收结构化的宠物健康监测报告
- 包含:宠物基本信息、监测时间段、各项行为指标统计、异常状况识别、健康评分、就医建议
- 准备视频输入
资源索引
- 必要脚本:见 scripts/pet_health_monitoring_analysis.py(用途:调用 API 进行宠物健康监测分析,本地文件使用 multipart/form-data 方式上传,网络 URL 由 API 服务自动下载)
- 配置文件:见 scripts/config.py(用途:配置 API 地址、默认参数和视频格式限制)
- 领域参考:见 references/api_doc.md(何时读取:需要了解 API 接口详细规范和错误码时)
注意事项
- 仅在需要时读取参考文档,保持上下文简洁
- 视频要求:支持 mp4/avi/mov 格式,最大 100MB,建议视频时长不少于 1 小时以反映日常行为规律
- API 密钥可选,如果通过参数传入则必须确保调用鉴权成功,否则忽略鉴权
- 分析结果仅供宠物健康参考,不能替代专业兽医诊断,发现异常请及时就医
- 禁止临时生成脚本,只能用技能本身的脚本
- 传入的网络地址参数,不需要下载本地,默认地址都是公网地址,api 服务会自动下载
- 当显示历史监测报告清单的时候,从数据 json 中提取字段 reportImageUrl 作为超链接地址,使用 Markdown 表格格式输出,包含"
报告名称"、"宠物类型"、"监测时间"、"点击查看"四列,其中"报告名称"列使用
宠物健康监测分析报告-{记录id}形式拼接, "点击查看"列使用[🔗 查看报告](reportImageUrl)格式的超链接,用户点击即可直接跳转到对应的完整报告页面。 - 表格输出示例:
报告名称 宠物类型 监测时间 点击查看 宠物健康监测分析报告-20260312172200001 猫 2026-03-12 17:22:00 🔗 查看报告
使用示例
# 分析本地监控视频(OpenClaw UI 上下文,使用 metadata id 作为 open-id)
python -m scripts.pet_health_monitoring_analysis --input /path/to/pet_monitor.mp4 --pet-type cat --monitor-days 1 --open-id openclaw-control-ui
# 分析网络监控视频(OpenClaw UI 上下文,使用 metadata id 作为 open-id)
python -m scripts.pet_health_monitoring_analysis --url https://example.com/pet_monitor.mp4 --pet-type cat --monitor-days 1 --open-id openclaw-control-ui
# 分析狗狗监控视频(OpenClaw UI 上下文,使用 metadata id 作为 open-id)
python -m scripts.pet_health_monitoring_analysis --input /path/to/dog_monitor.mp4 --pet-type dog --monitor-days 3 --open-id openclaw-control-ui
# 显示历史监测报告/显示监测报告清单列表/显示历史宠物健康报告(自动触发关键词:查看历史检测报告、历史报告、监测报告清单等)
python -m scripts.pet_health_monitoring_analysis --list --open-id openclaw-control-ui
# 输出精简报告
python -m scripts.pet_health_monitoring_analysis --input video.mp4 --pet-type cat --open-id your-open-id --detail basic
# 保存结果到文件
python -m scripts.pet_health_monitoring_analysis --input video.mp4 --pet-type cat --open-id your-open-id --output result.json
Files
31 totalSelect a file
Select a file to preview.
Comments
Loading comments…
