figma-build
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides JavaScript code snippets in
MEASURE.mdintended for execution within browser developer tools or automation environments (like Playwright). These scripts extract layout and style information (bounding boxes, padding, colors, etc.) to facilitate pixel-perfect UI comparisons. This execution is scoped to the component under test and is necessary for the skill's objective.\n- [INDIRECT_PROMPT_INJECTION]: The skill defines a process that ingests external data from a rendered UI. While this creates a potential surface for indirect injection if the UI content is untrusted, the risk is inherent to UI testing tools.\n - Ingestion points: Browser DOM metadata extracted via
document.querySelectorAllinMEASURE.md.\n - Boundary markers: None identified in the reporting scripts.\n
- Capability inventory: Agent-driven UI code modification and browser automation.\n
- Sanitization: No specific filtering or escaping of the extracted DOM values is mentioned before they are processed by the agent.
Audit Metadata