insurance-agent-activity-planner

Pass

Audited by Gen Agent Trust Hub on Jun 29, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches official libraries agentscope and dashscope from standard registries. These are recognized as legitimate vendor resources from the author.
  • [COMMAND_EXECUTION]: Includes a Python script scripts/validate_input.py for input validation. The script is benign and uses standard JSON processing without arbitrary execution.
  • [CREDENTIALS_UNSAFE]: Uses .env.example with placeholders for secret management, which is a standard and safe practice.
  • [SAFE]: The skill instructions define clear operational boundaries and compliance constraints. No signs of prompt injection, data exfiltration, or obfuscation were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 29, 2026, 06:40 AM
Security Audit — agent-trust-hub — insurance-agent-activity-planner