file-cleanup
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely focused on design file organization within Figma. It does not perform any network operations, access local system files, or execute external code. All actions are scoped to the Figma design environment.
- [SAFE]: The skill explicitly instructs the agent never to delete design content during cleanup. Deletion is defined as a separate, human-confirmed step, reducing the risk of accidental data loss.
- [SAFE]: The skill includes 'Figma Execution Limits' that prevent the agent from claiming access to private libraries, admin settings, or analytics, adhering to the principle of least privilege.
- [SAFE]: Indirect prompt injection is a theoretical risk as the skill processes Figma layer names and comments provided by users. However, the instructions mandate previews and change logs, ensuring that any anomalous behavior would be visible to the user before or after execution.
Audit Metadata