Sleep Quality Analysis Skill | 睡眠质量分析技能

This skill appears to perform sleep-video analysis, but contains behaviours you should verify before installing or running with real data: - Review where videos and generated reports will be stored locally (attachments/ and workspace data folder). If you are processing sensitive camera footage, plan secure storage or delete files after use. - The runtime asks the agent to read config files, including a workspace-level path (${OPENCLAW_WORKSPACE}/skills/smyx_common/scripts/config.yaml). Confirm that file does not contain secrets you don't want the skill to read, and understand that the skill may access workspace-level configuration. - The code includes a local SQLite DAO and writes to data directories — expect local persistence of metadata and possibly results. - The package bundles a large common library and an unrelated face-analysis skill; audit those files if you need assurance they do only expected network calls. In particular check ApiEnum.BASE_URL_* values in smyx_common config.yaml (they reference lifeemergence.com and dev/test hosts in examples). - The SKILL.md contains contradictory constraints (forbidding local memory reads while instructing local config reads and saving uploads). Ask the author or maintainer to clarify which local files the skill may access and to reduce ambiguity. Practical steps: run the skill in an isolated environment (sandbox/container), inspect/override config.yaml values to point to trusted API endpoints, avoid providing global workspace credentials, and delete stored attachments after testing. If you cannot validate the upstream API endpoints and config usage, treat the skill as untrusted for sensitive data.