figma-data-handling

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external, user-generated content through the Figma Comments API.\n
  • Ingestion points: Comments are retrieved from api.figma.com/v1/files/${fileKey}/comments as described in SKILL.md and references/comments-api.md.\n
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands are included in the provided API handling examples.\n
  • Capability inventory: The skill uses fetch for network requests and includes logic for database interactions (db.figmaCache.deleteMany) and console logging.\n
  • Sanitization: The skill includes high-quality redaction logic (safeFigmaLog, redactFigmaUser) to protect PII, although it does not specifically scan for prompt injection patterns within comment text strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:46 AM
Security Audit — agent-trust-hub — figma-data-handling