The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to perform directory checks using ls and execute several local validation scripts via node (e.g., measure_render.mjs, verify_states.mjs, axe_audit.mjs). These are used to ensure the generated dashboard meets accessibility and responsiveness standards.
[DYNAMIC_EXECUTION]: The skill references environment setup tools using npx (e.g., npx skills add, npx ux-ui-agent-skills init) and executes local JavaScript files at runtime to perform UI audits.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes user-provided briefs to generate dashboard code.
Ingestion points: User-supplied dashboard requirements (briefs) and data files from the local kit.
Boundary markers: The skill includes explicit instructions to stop and notify the user if the local kit is missing, preventing the agent from guessing file contents.
Capability inventory: Includes the ability to perform directory listings and execute local Node.js audit scripts.
Sanitization: There is no explicit sanitization described for the user-provided layout or data requirements, though the output is restricted to SVG and HTML structures.

data-dashboard