shoplazza-discounts

Pass

Audited by Gen Agent Trust Hub on Oct 29, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as a natural language interface to the Shoplazza CLI, creating a potential surface for indirect prompt injection where crafted user input might attempt to manipulate command generation.
  • Ingestion points: Natural language user intents for creating, updating, or deleting discounts (e.g., '满X减Y', 'Create a 15% off coupon') as described in the trigger phrase mapping.
  • Boundary markers: The skill relies on external safety protocols defined in a common skill file and lacks internal input delimiters or specific "ignore instruction" warnings within the prompt interpolation.
  • Capability inventory: The skill enables the execution of CLI commands with financial implications, such as creating discount codes, coupon campaigns, and flash sales, as well as lifecycle operations like cancel and delete.
  • Sanitization: The instructions explicitly forbid the agent from fabricating values and mandate the use of AskUserQuestion for missing required fields, while also encouraging the use of --dry-run for verification.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 29, 2026, 03:07 AM