Arrhythmia Early Warning Analysis Tool | 心律失常早期预警分析工具

Based on facial video, identifies abnormal rhythms such as premature beats, atrial fibrillation, tachycardia/bradycardia, assists in early detection of heart...

This skill appears to implement what it claims (sending facial video to a backend for arrhythmia-risk analysis) but has several privacy and persistence trade-offs you should consider before installing: - Where do the videos and analysis results go? The code posts data to external API base URLs configured in smyx_common (example: lifeemergence.com). Confirm the operator, privacy policy, data retention, and deletion procedures before uploading sensitive videos. - Local writes: the skill will save uploaded attachments and create/use a local SQLite DB under the workspace/data path and may create config.yaml files. Expect persistent artifacts on disk; if you need to remove them you should know their locations (OPENCLAW_WORKSPACE/data and the skill directory). - Credentials/open-id handling: the skill mandates an 'open-id' to record/query reports and will read it from config files or environment variables if present. Do not supply highly sensitive credentials as 'open-id'; ask the skill author how open-id is used and whether it's stored/transmitted. - Large dependency surface: the shared smyx_common requirements are extensive. If the platform would install these, consider the attack surface and prefer running in an isolated environment. - If you want to proceed: request from the publisher (or require) the following before trusting the skill: explicit API endpoints (full domain and operator), a privacy/data-retention statement, a list of exactly which dependencies are needed at runtime, and documentation of where files are stored and how to delete them. If any of those are missing or the backend operator is unknown, run the skill only in a sandbox and avoid uploading real-person videos until you can verify privacy controls.

Type: OpenClaw Skill Name: smyx-arrhythmia-early-warning-analysis Version: 1.0.1 The skill bundle provides a health analysis tool for detecting arrhythmia via facial video processing. It utilizes a structured framework with a shared utility library (smyx_common) that manages API communication with 'lifeemergence.com' and persists session tokens in a local SQLite database (smyx-common-claw.db). While SKILL.md contains 'Mandatory Rules' that strictly override the AI agent's default memory usage to force cloud API synchronization, these instructions are aligned with the functional requirement of maintaining consistent historical reports and do not exhibit malicious intent or unauthorized data exfiltration.