shoplazza-products
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest untrusted data from user utterances (e.g., product titles, spec values, tags, and review content) and use it to construct CLI commands that perform write operations on a Shoplazza store catalog.
- Ingestion points: Untrusted data enters the agent context through various user-supplied strings documented in
SKILL.md(e.g.,--title,--keyword,--vendor,--addfor tags, and spec values for+set-variants) and reference files likereferences/comments.md(review content) andreferences/gift-cards.md(gift card codes). - Boundary markers: The skill partially mitigates risk by instructing the agent to use
--dry-runand seek explicit user confirmation before executing destructive operations (like deletions or disabling gift cards) or "money operations" (like gift card creation). - Capability inventory: The skill possesses significant capabilities via the
shoplazza productsCLI, including creating, updating, and deleting products, variants, collections, gift cards, and customer reviews. - Sanitization: The instructions do not specify any sanitization, escaping, or validation of user-provided strings before they are interpolated into CLI command arguments or JSON data payloads, leaving the surface exposed to potential injection attacks embedded in the user's data.
Audit Metadata