product-price-research

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external marketplaces (Amazon, eBay, Shopify, and Shop.app) via the Crawlora API. This content, particularly product reviews and page descriptions, could contain adversarial instructions intended to influence the agent's behavior.
  • Ingestion points: Marketplace product details, seller feedback, and storefront page content are fetched via scripts/crawlora.sh and returned to the agent.
  • Boundary markers: The instructions do not define explicit delimiters (like XML tags or markdown blocks) to separate untrusted marketplace data from the agent's system instructions.
  • Capability inventory: The skill possesses the capability to execute network requests using the provided scripts/crawlora.sh helper.
  • Sanitization: The API documentation in reference/endpoints.md indicates that certain fields (like Shopify page bodies) are returned as "cleaned text only," which serves as a basic mitigation but does not fully prevent the risk of instruction-laden text being processed by the LLM.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 06:10 PM
Security Audit — agent-trust-hub — product-price-research