figma-implement-design
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is authored by a trusted organization (OpenAI) and utilizes official Figma service endpoints (mcp.figma.com) for design synchronization.- [INDIRECT_PROMPT_INJECTION]: The skill processes external design data from Figma, which introduces a potential surface for indirect prompt injection.
- Ingestion points: Figma node data and metadata fetched through get_design_context and get_metadata tools.
- Boundary markers: The skill instructs the agent to map design data to existing project conventions and design system tokens, providing a functional boundary for the code generation process.
- Capability inventory: Tool access is limited to fetching design information; no dangerous capabilities such as arbitrary shell execution, unauthorized file modifications, or external network exfiltration are present.
- Sanitization: No explicit sanitization of node content is performed, as the agent is expected to interpret design attributes for code translation.
Audit Metadata