figma-migration-deep-dive

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill interacts with the official Figma API domain (api.figma.com), which is a well-known and trusted service for design data management.
  • [SAFE]: Authentication is handled via environment variables (FIGMA_PAT), ensuring that sensitive API tokens are not hardcoded or exposed within the skill code.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external design data from Figma, creating an ingestion surface for untrusted content.
  • Ingestion points: Data is fetched from Figma API endpoints in the inventory and extraction functions.
  • Boundary markers: No specific delimiters are used to wrap the processed API data.
  • Capability inventory: The skill uses network operations (fetch) and local script execution (node) to perform its functions.
  • Sanitization: Implements basic string replacement for naming conventions but does not include sanitization for potential instructions embedded in design metadata.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:47 AM
Security Audit — agent-trust-hub — figma-migration-deep-dive