find-orders

Pass

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted content from external Amazon order pages (such as product titles or seller information).
  • Ingestion points: The skill instructions in references/order-workflow.md specify reading Amazon order history and detail pages.
  • Boundary markers: There are no explicit instructions or delimiters to wrap or sanitize content retrieved from Amazon to prevent the agent from following instructions embedded in that content.
  • Capability inventory: The skill summarizes and processes data from the browser to present findings to the user (SKILL.md).
  • Sanitization: No sanitization or validation of the text content retrieved from the web pages is performed before it is processed by the AI.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 20, 2026, 03:14 AM
Security Audit — agent-trust-hub — find-orders