viewports
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes external web pages and incorporates element metadata and text content into its analysis reports.
- Ingestion points: The
overflow.jsscript extracts tag names, CSS classes, element IDs, and text content (up to 40 characters) from DOM elements in the browser. - Boundary markers: The instructions do not specify delimiters or warnings for the agent when processing the extracted web content.
- Capability inventory: The skill uses
page.evaluateto execute its measurement script and captures screenshots of the rendered pages via the agent's browser tools. - Sanitization: Extracted text content is normalized for whitespace and truncated to a maximum of 40 characters.
Audit Metadata