Skills
skills/zai-org/glm-skills/glmv-prompt-gen/Gen Agent Trust Hub

glmv-prompt-gen

Pass

Audited by Gen Agent Trust Hub on Apr 16, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The script communicates with the official Zhipu AI API endpoint at open.bigmodel.cn to send visual data for analysis and receive generated prompts.
  • [CREDENTIALS_UNSAFE]: The skill requires a ZHIPU_API_KEY, which is securely handled through environment variables or user-provided arguments, following standard development practices.
  • [COMMAND_EXECUTION]: The skill includes a Python script that uses the requests library to interact with external APIs. No arbitrary or dangerous command execution was identified.
  • [DATA_EXFILTRATION]: Data transmission is limited to the user-provided images/videos and the API key sent to the official service provider (Zhipu AI) for the intended purpose of the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 16, 2026, 02:44 AM