← 返回 Skills 市场
Pet Soothing Trigger Analysis Skill | 宠物安抚触发分析技能
作者
smyx-skills
· GitHub ↗
· v1.0.0
· MIT-0
69
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install smyx-pet-calming-trigger-analysis
功能描述
Automatically triggers soothing mechanisms (playing relaxing sounds, activating laser toys) when pet anxiety, howling, or prolonged loneliness is detected; a...
安全使用建议
What to consider before installing/running this skill:
- Code and data flow: The skill will upload video files and query remote APIs (configured in smyx_common config.yaml). If you run it, your video(s) and user open-id may be sent to external servers. Review the RequestUtil / api_service code to confirm which endpoints and headers are used and what is transmitted.
- Local persistence: The package contains a SQLite DAO and will create files under your workspace/data and save uploaded attachments in the skill directory. If you are concerned about local storage of sensitive recordings, run in an isolated environment or inspect and control the attachments/db paths first.
- Undeclared env/config dependencies: Although the registry metadata lists no env vars, the code reads OPENCLAW_WORKSPACE, OPENCLAW_SENDER_OPEN_ID, FEISHU_OPEN_ID and skill/workspace config.yaml files. Do not provide sensitive credentials (API keys, tokens, or personal IDs) until you confirm exactly where they are sent.
- Mismatched components: The bundle includes an unrelated face_analysis (human health/TCM) module and a large shared library. That increases the runtime dependencies and surface for data collection beyond pet use-cases. Ask the publisher why these components are bundled and whether they are necessary.
- Practical steps: If you want to test safely, review skills/smyx_common/scripts/util.py and scripts/api_service.py to confirm network behavior and what fields are sent; run the tool in a restricted environment (no network access) first; or point api-url to a local mock endpoint to observe payloads. If you must run against the provider, avoid using production credentials or highly sensitive identifiers as the open-id until you trust the endpoint and have a privacy policy/owner contact.
- Missing provenance: There is no homepage or publisher details — lack of source attribution reduces trust. Consider requesting the author's identity, a hosted API specification, and a privacy statement before providing real user data or API keys.
功能分析
Type: OpenClaw Skill
Name: smyx-pet-calming-trigger-analysis
Version: 1.0.0
The skill bundle exhibits several high-risk behaviors and unusual architectural patterns. The 'SKILL.md' file contains 'Mandatory Memory Rules' that explicitly instruct the AI agent to ignore local memory files and LanceDB, forcing all data retrieval through a remote API (lifeemergence.com), which could be used to bypass local auditing. The 'smyx_common/scripts/util.py' file implements a 'silent registration' mechanism that automatically sends user identifiers—which the instructions suggest should be a phone number or username—to a remote endpoint. Additionally, 'smyx_common/scripts/skill.py' includes a 'subprocess.run' call to execute system commands via the 'openclaw' CLI, a high-risk capability. The bundle also contains the complete source code for an entirely unrelated 'face_analysis' skill, suggesting poor supply chain hygiene or hidden functionality.
能力标签
能力评估
Purpose & Capability
The skill description is a pet soothing/trigger feature, but the bundle includes a large shared library (skills/smyx_common) and a separate face_analysis skill (including TCM/health-analysis code). Those components introduce DB/DAO, many third-party dependencies, and endpoints for general AI analysis and health reports not strictly necessary for a narrow 'pet calming trigger' feature. This breadth is disproportionate to the stated simple purpose and increases the attack surface and data-handling scope.
Instruction Scope
SKILL.md imposes strict rules (forbids reading local memory files and LanceDB, mandates obtaining an open-id via specific config.yaml files or user input) and instructs the agent to save uploaded attachments to an attachments directory. The code, however, reads config files (skills/smyx_common/scripts/config.yaml), uses ConstantEnum.init which reads environment variables, and includes a local DAO/SQLite implementation that will create and write database files. Those actions contradict the 'forbid local memory' rule and broaden what the skill reads/writes at runtime. The skill also mandates calling cloud APIs for history and will upload videos (multipart/form-data) to external endpoints — that is expected for remote analysis but must be considered part of the scope.
Install Mechanism
No install spec (instruction-only) reduces install-time risk (nothing is automatically downloaded during installation). However, the included smyx_common/requirements.txt lists many packages and the codebase expects to run (imports requests, sqlalchemy, etc.). Although there is no remote download/install step in the manifest, running the skill in practice will require installing substantial third-party packages, which increases runtime dependency and maintenance risk.
Credentials
The skill declares no required env vars in metadata, but its code uses environment/config values: ConstantEnum.init reads OPENCLAW_SENDER_OPEN_ID, OPENCLAW_SENDER_USERNAME, FEISHU_OPEN_ID and Dao.get_db_path consults OPENCLAW_WORKSPACE. SKILL.md also mandates obtaining an 'open-id' from skill-local or workspace config files. The code also relies on external API base URLs configured in smyx_common/scripts/config.yaml (example base-url-open-api: https://open.lifeemergence.com/smyx-open-api). Asking the user to provide open-id and optionally API keys is plausible, but the skill silently depends on other envs/configs that were not declared in the top-level requirements, which is inconsistent and risky for secrets/exposure.
Persistence & Privilege
The code will create and write local artifacts: it saves uploaded attachments into the skill directory attachments and creates an SQLite DB under the workspace data directory (smyx-common-claw.db) via Dao.get_db_path. It also reads skill/workspace config YAML files. While it does not request 'always' privilege, it does persist data on disk and may read workspace-level config (i.e., other skills' config.yaml), which grants it access to some cross-skill configuration and persistent storage that a minimal analysis-only skill might not need.
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-calming-trigger-analysis - 安装完成后，直接呼叫该 Skill 的名称或使用
/smyx-pet-calming-trigger-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 pet-calming-trigger-analysis skill.
- Detects pet anxiety, howling, or extended loneliness from video and automatically triggers calming responses (e.g., soothing sounds, laser toys).
- Supports querying and displaying history of calming interventions via cloud API—local or long-term memory is strictly forbidden.
- Requires strict open-id verification before analysis; guides users to provide if not found.
- Delivers analysis and history reports in structured Markdown tables, with direct report links.
- Provides step-by-step setup, usage instructions, and constraints for accurate operation.
元数据
常见问题
Pet Soothing Trigger Analysis Skill | 宠物安抚触发分析技能 是什么？
Automatically triggers soothing mechanisms (playing relaxing sounds, activating laser toys) when pet anxiety, howling, or prolonged loneliness is detected; a... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 69 次。
如何安装 Pet Soothing Trigger Analysis Skill | 宠物安抚触发分析技能？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install smyx-pet-calming-trigger-analysis」即可一键安装，无需额外配置。
Pet Soothing Trigger Analysis Skill | 宠物安抚触发分析技能 是免费的吗？
是的，Pet Soothing Trigger Analysis Skill | 宠物安抚触发分析技能 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Pet Soothing Trigger Analysis Skill | 宠物安抚触发分析技能 支持哪些平台？
Pet Soothing Trigger Analysis Skill | 宠物安抚触发分析技能 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 Pet Soothing Trigger Analysis Skill | 宠物安抚触发分析技能？
由 smyx-skills（@18072937735）开发并维护，当前版本 v1.0.0。
推荐 Skills