What problem does it solve? Understanding a CARLA map's road layout normally requires writing ad-hoc Python against the Map/Waypoint API, and resolving phrases like "the 4-way junction in the middle" into a specific junction id is tedious guesswork. This Skill turns the loaded map into structured data — stats, junction lists with centre/bearing/arm counts, lane facts — so an agent can answer natural-language map questions and visually confirm the modelled topology matches the rendered one. ## Core Features & Use Cases - Map rundown: summary produces road/lane counts, extent, junction breakdown, density classification, and a one-paragraph prose description ready to relay. - Topology and junction visualization: topology --draw and junctions --draw overlay the road graph and labelled junction boxes via world.debug to verify the model against the rendered map. - Natural-language element resolution: junctions lists every junction with arm count, distance-to-centre, and bearing so phrases like "the northern 4-way junction" resolve to a concrete id. - Point queries and navigation: waypoint --at returns lane facts at a coordinate; navigate walks next() along a lane with optional drawing. - Use Case: A user asks "highlight the 4-way junction in the middle of Town10" — run junctions --arms 4, pick the smallest distance-to-centre, then draw it to confirm. ## Quick Start Ask the agent to give you a rundown of the currently loaded CARLA map's topology, or to find and highlight a specific junction such as the 4-way one nearest the centre.
