moai-domain-figma
Pass
Audited by Gen Agent Trust Hub on Mar 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
Bashtool and Node.jsfsmodule to interact with the local file system, enabling the creation of directories and the writing of design tokens and component code to disk. - [EXTERNAL_DOWNLOADS]: The skill retrieves design assets, metadata, and variables from Figma's official API using MCP tools such as
mcp__figma__get_design_contextandmcp__figma__get_variable_defs. - [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface.
- Ingestion points: Untrusted design data is ingested from external Figma files via MCP tools.
- Boundary markers: The implementation lacks specific delimiters or instructions to disregard potential instructions embedded in design data.
- Capability inventory: The skill can write executable code files and configuration files to the local environment.
- Sanitization: The skill writes content such as
generatedCodedirectly to files without performing validation or sanitization on the content received from the external API.
Audit Metadata