spree-storefront
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes architecture and provides code snippets that ingest data from the Spree Store API (e.g., product details, cart information), which creates a potential surface for indirect prompt injection attacks.
- Ingestion points: Data is retrieved via
@spree/sdkcalls such asspree.products.list()andspree.products.get()as documented inSKILL.md. - Boundary markers: The example React components (e.g.,
ProductPage) do not demonstrate the use of explicit delimiters or warnings to ignore instructions that might be embedded in product metadata. - Capability inventory: The skill documentation focuses on network operations to the Spree backend and client-side rendering; no high-privilege operations such as local file system writes or subprocess execution are present in the provided examples.
- Sanitization: No sanitization or validation of the text content fetched from the API is shown before it is rendered to the user interface in the provided examples.
Audit Metadata