figma-pixel-verify
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources that could contain malicious instructions designed to influence agent behavior.
- Ingestion points: The skill takes a Figma URL and a Runnable UI URL as inputs to extract specifications and measure styles (SKILL.md).
- Boundary markers: Absent; there are no instructions to the agent to use specific delimiters or ignore instructions found within the design specs or UI content.
- Capability inventory: The skill is granted the capability to execute code in the browser ('JS evaluate in page') and modify project files ('Apply fixes') based on findings (SKILL.md).
- Sanitization: Absent; no validation or sanitization of the external metrics or metadata is documented.
- [DYNAMIC_EXECUTION]: The skill performs runtime execution of JavaScript to inspect and extract UI properties.
- The 'Tool intent' section in SKILL.md explicitly allows the use of 'JS evaluate in page' for reading computed styles and box metrics.
- The reference.md file provides a JavaScript template intended for execution at runtime to query the DOM and retrieve layout and style information.
Audit Metadata