tui-action-sheet
Pass
Audited by Gen Agent Trust Hub on Apr 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's primary function is to guide the AI in rendering ASCII-based Action Sheet components and generating metadata for drawing tools. No malicious patterns such as credential theft, remote code execution, or unauthorized network access were detected in the instructions or examples. All operations are confined to the AI's generation context.- [PROMPT_INJECTION]: Analysis of the input processing logic identifies a surface for indirect prompt injection, as user-provided data from the input model JSON is interpolated into structured output blocks (PENCIL_BATCH_DESIGN) without explicit sanitization instructions. This is a common architectural pattern in design-to-code skills and is assessed as safe within the scope of its intended functionality.
- Ingestion points: Untrusted data enters via the JSON model (props, state, style, etc.) defined in the workflow of SKILL.md.
- Boundary markers: None present; the skill lacks specific delimiters or warnings to ignore instructions embedded in the processed data.
- Capability inventory: The skill generates instructions for an external drawing tool (Pencil MCP) but has no autonomous execution capabilities of its own.
- Sanitization: No escaping, validation, or filtering of input string content is required by the instructions before interpolation into code-like output blocks.
Audit Metadata