glmocr-formula
Pass
Audited by Gen Agent Trust Hub on Mar 22, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill communicates with the official ZhiPu AI layout parsing endpoint (
https://open.bigmodel.cn/api/paas/v4/layout_parsing). This is a well-known service and the endpoint is hardcoded to prevent data redirection to unauthorized servers. - [COMMAND_EXECUTION]: The agent executes a local Python CLI script (
scripts/glm_ocr_cli.py) to process images and PDFs. The script's logic is transparent, single-purpose, and does not perform any suspicious system-level operations. - [CREDENTIALS_UNSAFE]: Authentication is handled securely through the
ZHIPU_API_KEYenvironment variable. The skill provides clear instructions for users to configure this safely and avoids hardcoding sensitive tokens.
Audit Metadata