design
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-defined screen requirements to generate structured files.
- Ingestion points: User-provided descriptions and feature lists (SKILL.md, Step 1).
- Boundary markers: Uses structured Markdown and YAML frontmatter to separate design content from instructions.
- Capability inventory: The skill performs file system write operations restricted to the
./design/directory and its subdirectories. - Sanitization: No explicit sanitization is described, but the skill defines a rigorous process for translating natural language into the specific Screen Tree Notation (STN) grammar.
Audit Metadata