tui-fab
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input data (UI properties, state, and labels) to generate structured specifications and TUI renderings. This creates an attack surface where malicious instructions could be embedded in the input data to influence the behavior of the agent or downstream tools (e.g., Pencil MCP).
- Ingestion points: UI property JSON model defined in
SKILL.md(e.g.,props,state,style). - Boundary markers: The instructions do not define specific delimiters or "ignore instructions" warnings for the processed input data.
- Capability inventory: The skill generates
PENCIL_BATCH_DESIGNcalls and component specifications intended for use with Pencil MCP drawing workflows. - Sanitization: There is no explicit requirement for sanitizing or validating input text before it is interpolated into the output blocks.
Audit Metadata