figma-use
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of instructional content and JavaScript code snippets focused on legitimate Figma automation tasks. The documentation provides clear best practices for interacting with the Figma Plugin API, including node creation, layout management, and design system integration. All patterns provided are consistent with standard plugin development workflows.\n- [PROMPT_INJECTION]: The skill provides patterns for reading node names, text content, and component descriptions from Figma canvases. While this exposes a potential indirect prompt injection surface where an attacker could place malicious instructions in a Figma file to influence agent behavior, it is an inherent characteristic of any tool designed to process external design metadata.\n
- Ingestion points: Metadata traversal and read-only patterns in
references/common-patterns.md,references/component-patterns.md, andreferences/validation-and-recovery.md.\n - Boundary markers: No explicit delimiters are used in the provided snippets to separate Figma content from tool instructions.\n
- Capability inventory: The
use_figmatool permits dynamic JavaScript execution within the Figma context as described inSKILL.md.\n - Sanitization: No specific filtering or escaping logic for external node data is included in the reference snippets.
Audit Metadata