create-structure
Fail
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: HIGHINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from a user-provided Markdown component specification. 1. Ingestion points: Input .md file parsed in Step 0. 2. Boundary markers: None; the skill lacks markers to differentiate data from instructions. 3. Capability inventory: Arbitrary JS execution via figma_execute and use_figma. 4. Sanitization: Absent; content is directly interpolated into JS payloads.
- [DYNAMIC_EXECUTION]: The skill generates JavaScript code at runtime by replacing placeholders in templates with data from the input spec. Because names and notes are not sanitized, a malicious Markdown file could inject arbitrary JavaScript into the Figma environment.
- [COMMAND_EXECUTION]: The skill's instructions to skip user confirmation and 'pick the most defensible option' allow for autonomous execution of these generated commands, significantly increasing the risk of unvetted actions being performed on the user's Figma account.
Recommendations
- AI detected serious security threats
Audit Metadata