TCM Facial Diagnosis Analysis Tool

Problem

When users have an MP4 video or a video URL containing a face and want a reproducible TCM facial diagnosis reference, manual inspection is hard to standardize. This skill turns video input, cloud analysis, structured reports, and history lookups into one script-driven workflow.

Capabilities and flow

• Input: Supports local MP4 paths or network video URLs; the face should be clearly visible, with a typical maximum size of 10MB.
• Identity: A valid open-id is required before analysis to store and query historical reports; values must not be guessed or generated.
• Analysis: Uses scripts/face_analysis.py to call the server-side diagnostic API and evaluates TCM inspection signals such as complexion, luster, and texture.
• Output: Returns structured results covering constitution, organ-system condition, health alerts, and wellness suggestions, with basic, standard, or json detail levels.
• History: --list queries the cloud API for prior reports and can present them as a Markdown table with report name, analysis time, and view link.

Boundaries and notes

The output is for health reference only and cannot replace professional medical diagnosis. Historical reports must come from the cloud interface rather than local memory files, and optional API keys must authenticate successfully when provided.