What problem does it solve? When a design request arrives in Craft/design mode without clear UI signals, it is unclear which specialized design workflow should handle it. This Skill resolves that ambiguity by classifying the user's intent and dispatching to the correct domain skill instead of improvising a procedure. ## Core Features & Use Cases - Intent Classification: Reads the user message, selected canvas node, and editor state to distinguish UI screens, slide decks, posters, livestream posters, and design-to-code requests. - Skill Dispatch: Loads exactly one matching domain skill (ardot-ui-design, ardot-slides, ardot-poster, livestream-poster, or ardot-design-to-code) and defers to it together with ardot-design-core. - Hard Rule Enforcement: Applies always-on canvas design rules, including no sub-agents, a three-part reply format, target-node priority, and screenshot verification via capture_screenshot. - Use Case: A user in design mode says "make a landing page for our product launch" — the router matches the UI/interface intent, loads ardot-ui-design, and the design work proceeds under that skill's workflow. ## Quick Start Describe the design you want on the canvas, such as asking to create a mobile app dashboard screen, and the router will classify it and load the right design skill.
