gongzhonghao-search

Pass

Audited by Gen Agent Trust Hub on Aug 11, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXFILTRATION]: The skill transmits search keywords to the doubaoya.com API. This is the core intended functionality of the skill and is limited to communicating with the service provider's infrastructure. No sensitive local data is accessed or transmitted.
  • [CREDENTIALS_UNSAFE]: Authentication is performed using the DOUBAOYA_API_KEY environment variable. The skill instructions include explicit constraints for the AI agent to prevent the leakage or display of this sensitive key to users.
  • [COMMAND_EXECUTION]: The skill executes a Python script (scripts/search_gzh.py) using standard libraries. The script is self-contained, does not download external code, and uses Python's standard urllib for secure network communication.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 11, 2026, 07:48 PM
Security Audit — agent-trust-hub — gongzhonghao-search