Mental Health Analysis Tool | 心理健康分析工具

Analyzes human mental health and psychological behavior, supports identifying common psychological problem tendencies through video analysis, and provides st...

What to consider before installing: - Data exfiltration & privacy: This skill uploads user video (local file or URL) to an external API host (configured via smyx_common config). That is expected for a cloud-based analysis service, but it means sensitive video will leave your environment — only use with explicit consent and with non-sensitive footage. Verify the API host (config.yaml values) and the service's privacy policy before sending personal data. - Open-id / API keys: The skill requires an open-id to run and accepts an optional api-key. The open-id retrieval flow reads local config.yaml files and environment variables; do not supply credentials you don't want associated with uploaded reports. The SKILL.md forbids fallback to local long-term memory, but the code does read/write config files and may persist a local SQLite DB — this is a mismatch to be aware of. - Local writes: The code can create config files and a sqlite DB under the workspace/data directory. If you want no disk persistence, do not install/run this skill in your normal environment; run it in an isolated/test environment instead. - Code provenance: The skill owner and homepage are unknown. If you need higher assurance, request the developer to provide: (1) an authoritative API host and privacy policy, (2) a short audit of RequestUtil (network targets and headers) showing exactly where data is sent, and (3) confirmation whether local DB/attachments will be created and where. Without that, treat this as untrusted third‑party software. - Practical tests: If you proceed, test with non-sensitive sample videos and inspect network traffic or RequestUtil to confirm the destination(s). Also inspect or run the skill in a throwaway container to see what files it creates (config.yaml, workspace/data/*.db). What would change this assessment: If the developer supplies an official homepage, privacy policy, and a clear statement of the API host(s) used (and a code-level review of RequestUtil showing no hidden endpoints), I would raise confidence toward benign. Conversely, if RequestUtil or any included module is found to POST data to unexpected third-party domains, the risk would increase.

Type: OpenClaw Skill Name: smyx-psychology-analysis Version: 1.0.0 The skill bundle provides a psychology and facial analysis tool that interacts with a cloud-based API (lifeemergence.com). It includes a robust framework (smyx_common) for handling authentication, local session caching via a SQLite database (smyx-common-claw.db), and structured reporting. While SKILL.md contains 'Mandatory Memory Rules' that instruct the AI agent to ignore local memory files in favor of cloud API data, these appear to be functional constraints designed to ensure data consistency and prevent hallucinations rather than malicious prompt injections. The code follows a standard tool-use pattern, validating inputs and communicating only with the developer's stated infrastructure (e.g., open.lifeemergence.com).