figma-automation
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted external data from Figma designs and comments, which presents a surface for indirect prompt injection.\n
- Ingestion points: Data enters the context from potentially attacker-controlled sources via
FIGMA_GET_FILE_JSONandFIGMA_GET_COMMENTS_IN_A_FILE.\n - Boundary markers: The instructions lack defined delimiters or specific warnings to ignore instructions embedded in retrieved Figma content.\n
- Capability inventory: The skill possesses write capabilities such as
FIGMA_ADD_A_COMMENT_TO_A_FILE, which could be abused if the agent follows malicious instructions found in the design data.\n - Sanitization: No validation or sanitization of retrieved Figma content is mentioned before processing.\n- [EXTERNAL_DOWNLOADS]: The skill directs users to configure an external MCP server at
https://rube.app/mcp. While this is part of the intended setup for this toolkit, it relies on a non-standard third-party domain.
Audit Metadata