batterdeals

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external sources, including a coupon API (grabcaramel.com), deal community websites (Reddit, Pepper, Slickdeals), and arbitrary online shop pages. * Ingestion points: The subagent fetches HTML and JSON data from these sources in Step 2. * Capability inventory: The agent has web fetching and browser automation (Playwright) capabilities. * Boundary markers: The instructions lack explicit boundary markers or 'ignore instructions' directives for the ingested data. * Sanitization: There is no evidence of sanitization or filtering of the content retrieved from external sources before it is processed by the agent.
  • [EXTERNAL_DOWNLOADS]: The skill makes network requests to grabcaramel.com to fetch coupon data and interacts with various third-party retail and aggregator websites to retrieve code information.
  • [COMMAND_EXECUTION]: The skill uses browser automation tools (Playwright) to interact with external websites, add items to guest carts, and test promo codes in real-time.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 01:40 AM