ecommerce-amazon-keyword-intelligence

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external search term data retrieved via API, which introduces a surface for indirect prompt injection.
  • Ingestion points: Data from the intelligentQuery API endpoint (referenced in references/api.md and scripts/aba_query.py) is written to local storage and presented to the agent.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded content are used when presenting retrieved search terms to the agent.
  • Capability inventory: The skill relies on a Python script (scripts/aba_query.py) and suggests using CLI tools like jq to process resulting data files.
  • Sanitization: The scripts/aba_query.py script saves the raw JSON response without sanitization or filtering of the search term text content.
  • [EXTERNAL_DOWNLOADS]: The skill makes network requests to the NexScope API infrastructure to retrieve data.
  • Communicates with the service endpoint defined by the NEXSCOPE_PROXY_BASE environment variable in scripts/aba_query.py.
  • This is a legitimate vendor resource used to provide the skill's primary functionality of querying keyword intelligence.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 04:13 AM
Security Audit — agent-trust-hub — ecommerce-amazon-keyword-intelligence