blueprint
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources such as support tickets, process documentation, and stakeholder interviews.
- Ingestion points: Technical architecture reviews and research evidence (SKILL.md).
- Boundary markers: No specific delimiters or safety instructions are defined for processing untrusted data.
- Capability inventory: Writing HTML files to the working directory and interacting with Figma/Pencil MCP tools.
- Sanitization: The skill mandates that generated HTML must be self-contained with no JavaScript, external CSS, or external fonts.
- [EXTERNAL_DOWNLOADS]: The skill integrates with Figma and pencil.dev, which are recognized well-known services for design and diagramming.
- [DYNAMIC_EXECUTION]: The skill dynamically generates HTML visualizations. The risk of script injection is significantly reduced by the explicit prohibition of JavaScript in the output code.
Audit Metadata