prototype-to-figma-skill
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes prototype source code which constitutes untrusted external data. This creates a surface where malicious instructions embedded in the prototype (such as comments or string literals) could attempt to influence the agent's behavior during the design generation process.\n
- Ingestion points: Prototype source files including React, Vue, Svelte, or HTML files processed in the analysis step.\n
- Boundary markers: The skill does not define clear delimiters or instructions to ignore embedded code-level prompts.\n
- Capability inventory: The skill uses a wide range of Figma MCP tools to create files, search design systems, and add annotations to canvas elements.\n
- Sanitization: There is no evidence of content sanitization or validation of the source code before it is interpreted for interaction flow mapping.\n- [COMMAND_EXECUTION]: The documentation provides instructions for the user to manually execute shell commands using
catandgrepon local configuration files (claude_desktop_config.jsonandmcp_config.json). While intended for installation verification, these files are part of the local environment's configuration.\n- [EXTERNAL_DOWNLOADS]: The skill references the use ofnpx @figma/mcp-server. As this originates from an official and well-known service provider (Figma), this download is considered safe and functional for the skill's purpose.
Audit Metadata