The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes Node.js to run an internal audit script (scripts/gui_design_audit.mjs) and Python 3 to run a layout validation tool from a separate skill (layout-overflow-guard). These executions are part of the core functionality for auditing design specifications and are restricted to predefined file paths.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external design briefs (Markdown) and GUI specifications (JSON). This processing constitutes a potential attack surface for indirect instructions.
Ingestion points: The skill consumes design briefs and JSON specifications as defined in its IO contract.
Boundary markers: No specific delimiters or instructions to ignore embedded content are provided to the agent when processing these inputs.
Capability inventory: The agent has access to Bash, Read, Write, Edit, Grep, and Glob tools.
Sanitization: The skill provides a deterministic validation layer via gui_design_audit.mjs, which performs type and range checking on the JSON specification.
[EXTERNAL_DOWNLOADS]: The instructions recommend installing or syncing the layout-overflow-guard skill from the platform's skill catalog if it is not already present. This represents a functional dependency within the agent environment rather than a download from an untrusted remote server.

beautiful-gui-design