enable-feedproof

Pass

Audited by Gen Agent Trust Hub on Oct 22, 2026

Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: Reports request metadata, such as the visit path and user-agent string, to the FeedProof service at feedproof.fly.dev. This behavior is the primary intended function of the skill and utilizes a well-known hosting service.
  • [INDIRECT_PROMPT_INJECTION]: The skill integrates Shopify metafield data into product descriptions and JSON-LD schema, creating a potential surface for indirect prompt injection if the source data in Shopify is untrusted. * Ingestion points: Retrieves feedproof.* metafields from the Shopify Storefront API within lib/shopify/fragments/product/index.ts. * Boundary markers: The skill applies escapeMarkdown in lib/markdown/product/index.ts to delimit product attributes within the generated documentation. * Capability inventory: The skill's functionality is restricted to authenticated metadata reporting and data transformation; it does not possess capabilities for arbitrary command execution, privilege escalation, or access to sensitive local files beyond environment variables. * Sanitization: External values are trimmed and escaped before being interpolated into markdown and JSON-LD templates to maintain structural integrity.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 22, 2026, 12:09 PM