amazon-product-detail

Pass

Audited by Gen Agent Trust Hub on Aug 4, 2026

Risk Level: SAFE
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill generates a JavaScript payload using a local Python script (scripts/extract-product-detail.py) which is then passed to the browser-act tool via shell command substitution. This is a standard and transparent pattern for browser-based extraction skills.
  • [INDIRECT_PROMPT_INJECTION]: The skill has a data ingestion surface as it scrapes product descriptions, reviews, and AI summaries from external Amazon pages (File: scripts/extract-product-detail.py). While it returns data in a structured JSON format, it lacks explicit sanitization or boundary markers to prevent the agent from potentially interpreting scraped text as instructions. The skill has capabilities to execute code via the browser-act tool.
  • [COMMAND_EXECUTION]: Employs shell command substitution ($(python scripts/extract-product-detail.py)) to feed generated logic into the browser automation tool. The script source is local and its behavior is deterministic and limited to extraction.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 4, 2026, 05:53 AM
Security Audit — agent-trust-hub — amazon-product-detail