High-risk tools in NWO Robotics
25 of the 61 tools in NWO Robotics are classified as high risk. This page profiles those tools specifically, with recommended policy actions and the attack patterns that target them.
Every operation listed below is an action PolicyLayer recommends controlling at the transport layer. Open any tool to see the full profile, risk score, and YAML policy snippet.
Tools at high risk
-
detect_objectsExecuteRun computer vision to detect objects
-
execute_robot_taskExecuteSend a Vision-Language-Action command to a robot
-
nwo_auto_calibrateExecuteAutomated calibration of joint offsets, camera extrinsics, and sensor bias. <5min calibration time.
-
nwo_connect_ros2_robotExecuteConnect NWO API to ROS2-enabled robot via WebSocket bridge. Supports UR5e, Panda, Spot, and generic robots.
-
nwo_create_rl_envExecuteCreate cloud-hosted RL environment for training robot policies. First cloud RL training system optimized for physical robots.
-
nwo_detect_gestureExecuteReal-time hand gesture recognition for intuitive robot control. Wave to stop, point to direct, draw paths.
-
nwo_edge_inferenceExecuteUltra-low latency inference via global edge network (200+ locations). <50ms response time worldwide.
-
nwo_emergency_stopExecuteTrigger emergency stop on robot. Immediate halt with guaranteed <10ms response.
-
nwo_enable_safety_monitoringExecuteEnable continuous real-time safety monitoring with dynamic speed adjustment.
-
nwo_execute_actionsExecuteExecute a sequence of robot actions with safety checks and progress tracking.
-
nwo_execute_behavior_treeExecuteExecute hierarchical task execution using Behavior Trees. Complex task orchestration with conditional logic and error recovery.
-
nwo_execute_subtaskExecuteExecute a specific subtask from a generated plan with progress tracking.
-
nwo_inferenceExecuteStandard vision-language-action (VLA) inference. Send camera images and natural language instructions to get robot actions.
-
nwo_inference_with_routerExecuteIntelligent model router inference. Automatically selects the best model for the task based on classification.
-
nwo_initialize_slamExecuteInitialize SLAM (Simultaneous Localization and Mapping) for persistent robot mapping. Creates a continuous spatial map of the environment that the robot remembers across sessions.
-
nwo_launch_playgroundExecuteLaunch interactive simulation playground to test agents before real deployment.
-
nwo_localizeExecuteLocalize robot within existing SLAM map. The robot remembers previously mapped areas and localizes without re-mapping.
-
nwo_mqtt_publishExecutePublish sensor data or commands via MQTT to the NWO IoT network. Supports 1000+ agents.
-
nwo_plan_motionExecuteMoveIt2 motion planning integration. Plan collision-free trajectories using RRT-Connect, OMPL, and CHOMP algorithms.
-
nwo_process_voice_commandExecuteVoice-activated robot control. Wake word detection and natural language command parsing.
-
nwo_send_ros2_actionExecuteSend action directly to ROS2 robot through the bridge.
-
nwo_simulate_trajectoryExecuteCloud-hosted physics simulation for trajectory validation. High-fidelity MuJoCo simulation.
-
nwo_task_plannerExecuteBreak down complex high-level instructions into executable subtasks. AI analyzes and creates ordered plan.
-
nwo_train_policyExecuteTrain robot policy using Stable Baselines 3 (SB3) integration. Includes PPO, SAC, and DDPG algorithms with Sim-to-Real transfer.
-
stop_robotExecuteEmergency stop a robot
Attacks that target this class
High-risk tools in any server share these documented attack patterns. Each links to the full case and the defensive policy.