tui-nvue
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided JSON data (specifically within the 'props', 'modelValue', and 'state' fields) and interpolates this content into structured output blocks such as 'TUI_RENDER' and 'PENCIL_BATCH_DESIGN'.
- Ingestion points: Input data processed via the 'Input Model' defined in SKILL.md.
- Boundary markers: The skill uses explicit markdown headers (e.g., '### OUTPUT: TUI_RENDER') to delimit generated content, which helps separate data from instructions but does not fully prevent injection.
- Capability inventory: The skill generates specifications and batch commands for the 'Pencil MCP' tool, which involves drawing and layout operations.
- Sanitization: No specific sanitization or escaping logic is defined for the input data before it is rendered into the output blocks.
- [SAFE]: The documentation references provided in the skill ('full-stack-skills/tui-sources/components.json') point to resources owned by the skill's author, which is a standard and expected practice for providing context without using external or untrusted domains.
Audit Metadata