amazon-product-finder

Pass

Audited by Gen Agent Trust Hub on Feb 14, 2026

Risk Level: LOW
Full Analysis
  • SAFE (LOW): A review of all scripts and metadata found no evidence of malicious patterns, such as command injection, data exfiltration, or obfuscation. The behavior matches the stated purpose.
  • Indirect Prompt Injection (LOW): The skill handles user-provided keywords but lacks exploitable capabilities.
  • Ingestion points: keyword argument in scripts/find_products.py sourced from user command line input.
  • Boundary markers: None present.
  • Capability inventory: Only includes standard output printing and local dictionary access. No file-writing, network access, or subprocess execution.
  • Sanitization: None.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 14, 2026, 09:35 PM