glmocr-formula
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill communicates with the official ZhiPu AI platform through a hardcoded API endpoint, which is a recognized and well-known service.
- [SAFE]: API credentials are managed securely through environment variables (ZHIPU_API_KEY), following industry-standard practices for secret management.
- [SAFE]: The Python implementation in scripts/glm_ocr_cli.py is transparent and strictly limited to processing input files and interacting with the authorized API.
- [SAFE]: The agent's instructions include mandatory restrictions that limit its actions to the provided script, reducing the potential for unauthorized tool usage or unexpected behavior.
Audit Metadata