figma-static-pass
Pass
Audited by Gen Agent Trust Hub on Jul 14, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The instructions are strictly task-related and do not attempt to bypass safety constraints or override model behavior. The skill processes external Figma data (ingestion point: get_design_context) to create files. While this creates a surface for indirect injection, the risk is mitigated by the skill's specific task scope and the absence of high-privilege commands (boundary markers: naming contract gate; capability inventory: file write, asset download; sanitization: naming validation).
- [DATA_EXFILTRATION]: No patterns of sensitive data exfiltration were detected. The skill handles Figma design data and assets within the local development environment.
- [EXTERNAL_DOWNLOADS]: Design assets are downloaded from Figma-provided URLs to the project's local directory, which is standard behavior for design-to-code tools.
- [COMMAND_EXECUTION]: The skill does not require the execution of shell commands or arbitrary code.
Audit Metadata