figma-use-figjam

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides reference documentation for interacting with the Figma Plugin API via MCP tools. All code examples follow standard API usage patterns. No malicious intent, obfuscation, or unauthorized access patterns were detected.\n- [INDIRECT_PROMPT_INJECTION]: The skill describes tools like get_figjam that ingest board data into the agent's context. While this creates a potential surface for indirect prompt injection if a board contains malicious text, the skill itself does not contain malicious instructions and focuses on legitimate board management tasks.\n
  • Ingestion points: FigJam node tree XML returned by the get_figjam tool.\n
  • Boundary markers: The skill does not explicitly provide delimiters or safety instructions for handling untrusted board content.\n
  • Capability inventory: Node creation and modification via use_figma (Figma Plugin API), asset uploads via upload_assets.\n
  • Sanitization: No explicit sanitization or validation of board content is defined in the documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 04:33 AM
Security Audit — agent-trust-hub — figma-use-figjam