design-to-motion
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [SAFE]: The skill demonstrates strong security posture regarding credential management. It explicitly warns against printing, echoing, or leaking Figma personal access tokens, instructing the agent to utilize secure storage such as the OS keychain or environment variables.- [COMMAND_EXECUTION]: The skill utilizes the
ffmpegcommand-line tool to process image frames into MP4 and GIF formats. This is a legitimate and necessary operation for the skill's primary function of exporting design animations.- [DYNAMIC_EXECUTION]: To achieve deterministic animation recording, the skill employs browser-based JavaScript execution (e.g.,page.evaluateandpage.clock). This use of dynamic execution is confined to the local animation stage and is required for precise frame-by-frame capture.- [INDIRECT_PROMPT_INJECTION]: The skill processes external design data from Figma, creating a theoretical attack surface for indirect injection via malicious layer names or metadata. However, the risk is mitigated as the skill interprets this data for layout and motion properties rather than executing it as natural language instructions. - Ingestion points: Design tree, bounds, and image data fetched from the Figma REST API (
api.figma.com). - Boundary markers: None; the data is treated as structural design configuration.
- Capability inventory: Local command execution for media encoding (
ffmpeg) and browser-based JavaScript execution (page.evaluate). - Sanitization: None specified, though properties are mapped to CSS/JS layout logic rather than prompt context.
Audit Metadata