What problem does it solve? Getting a vehicle in a CARLA simulation to a specific location requires wiring up route planning, agent control loops, and sync-mode ticking by hand. This Skill packages that procedure so an agent can drive a car to a destination or plan a route without improvising against the PythonAPI. ## Core Features & Use Cases - Autonomous point-to-point driving: The go command drives the ego vehicle to a destination using BasicAgent, BehaviorAgent, or ConstantVelocityAgent, handling obstacle avoidance and traffic-light obedience. - Route planning without driving: The route command uses GlobalRoutePlanner to compute a waypoint path between two points, summarize the maneuver sequence, and optionally draw it in the world. - Driving style and rule control: Choose cautious/aggressive behavior, set target speed, or ignore traffic lights, vehicles, and stop signs for scenarios like emergency runs. - Use Case: A user asks to "drive the ego to the 4-way junction" — the Skill resolves the junction center via map-waypoints, then runs go --to=<center> and reports arrival or timeout. ## Quick Start Ask the agent to drive the ego vehicle to coordinates (-24, -57, 1) in the running CARLA server using the navigate-to skill.