Isaacsim MCP Server

Generate a 3D model from text or image using Beaver3D, then load it into the scene. Args: text_prompt: Text description for 3D generation. image_url: URL of an image for 3D generation. position: [x, y, z] world position for the generated model. scale: [sx, sy, sz] scale factors.

Diagnostic tool: recent WARN/ERROR logs plus captured print() output. Captures carb.log_*/omni.log WARN+ERROR and stdout from execute_script / reload_script (tagged [PRINT]). Plain print() outside those captured contexts may not appear. Defaults are agent-friendly: non-destructive (clear=False) and scoped to the current run (since_last_play=True) so you see logs from what you just did, not stale entries from previous runs. Args: clear: If True, empty the buffer after reading. Default False. count: Maximum number of log entries to return. since_last_play: If True (default), retur…

Diagnostic tool: get joint drive configuration for a robot articulation. Returns stiffness, damping, limits, target vs actual positions, and position error for each joint. Call this when: - Joint drives are not tracking targets (check position_error) - Joints are oscillating or unstable (check stiffness/damping ratio) - Joints hit limits unexpectedly (check lower_limit/upper_limit) Units: gains are per-radian (angular) or per-meter (linear), per each joint's `gain_units`. Write them back with set_drive_params, which takes the same units. Args: prim_path: USD path to the robot articulati…

Read current joint positions from a robot. Units: radians for revolute joints, meters for prismatic joints. Joint order matches the joint_names from get_robot_info. For a combined step-and-read, prefer step_simulation with observe_joints. Args: prim_path: The prim path of the robot.

Get point cloud data from a lidar sensor. Requires the timeline to be playing — RTX lidar data is produced by Replicator while the sim runs, and a sweep only completes on some frames, so an empty read means "not this frame", not "saw nothing". By default returns a summary rather than the raw cloud: point_count, bounds, and the nearest hit. A full sweep is tens of thousands of points and megabytes of JSON, which is rarely what you want in a response. Args: prim_path: Prim path of the lidar sensor. max_points: Include this many points in the response, sampled at an even stride…

Diagnostic tool: get physics state for a prim. Returns rigid body status, velocities, kinematic flag, and collision info. `mass` is included only when the prim carries a UsdPhysics MassAPI — objects created by create_object do not, and take their mass from the collider's density. Velocity units: linear_velocity in m/s, angular_velocity in rad/s. Velocities are only non-zero once the simulation has advanced — step the simulation (or play) before reading them. Call this when: - Objects fall through the ground (check collision enabled) - Objects don't move when expected (check is_kinematic, mass…

Get detailed information about a specific prim. Returns type, children, and a transform block. There is no bare "position": position_local — parent-relative, the value transform_object writes. position_world — where the prim actually is on the stage. Use this for distance, reach or contact; for a nested prim such as a robot link the two differ by the parent's pose. position_world_source is "usd" (from the authored transform) or "physics" (measured, on Newton). On Newton, position_local_warning means only position_local is the spawn pose (position_worl…

Get robot joint information including names, DOF count, joint types, and limits. Call this after create_robot to understand the robot's kinematic structure. Returns joint names ordered by DOF index, joint types (revolute/prismatic), and joint limits (radians for revolute, meters for prismatic — each entry carries its own `units`). Args: prim_path: The prim path of the robot.

Ping the Isaac Sim extension server and return scene information including stage path, assets root, and prim count.

Get the current simulation state: timeline status (playing/stopped/paused), simulation time, and physics dt. step_simulation does NOT require a running timeline — do not play just to step.

Import a robot from a URDF file into the scene. Args: urdf_path: Path to the URDF file. prim_path: Prim path for the imported robot. position: [x, y, z] world position.

List all available robots discovered from the Isaac Sim asset server. Returns robot keys, descriptions, manufacturers, and asset paths. The list is auto-discovered at startup and reflects the actual assets available in your Isaac Sim version.

List all available environments discovered from the Isaac Sim asset server. Includes warehouses, offices, outdoor scenes, and more.

List the prims directly under root_path, optionally filtered by type. One level deep by default, so `list_prims("/")` names /World and /Environment rather than everything inside them — a robot alone is hundreds of prims. The response echoes `recursive` so a shallow answer is never mistaken for a complete one. Pass recursive=True to walk the whole subtree. That is the one you want when checking whether something was really deleted, or when hunting a prim nested under a robot: a Camera at /World/Arm/EyeCam does not appear in a shallow listing of /World. Args: root_path: Root path to start…

Load a pre-built environment into the scene. Supports fuzzy matching. Call list_environments first for options, and call this BEFORE create_physics_scene — the reverse order leaves two collision floors, and is reported as "collision_floor_warning". Y-up and/or centimetre environments are rotated and rescaled to match the stage; "corrections" reports what was applied. Read prim_path from the response rather than assuming it. "bounds" carries two heights — use the right one: floor_height — the surface objects rest on. Place with position=[x, y, floor_height]. bounds_min…

Load a USD asset from a URL or file path into the scene. Args: usd_url: URL or local path to the USD file. prim_path: Prim path for the loaded asset. position: [x, y, z] world position. scale: [sx, sy, sz] scale factors.

Modify properties of an existing light. Args: prim_path: The prim path of the light to modify. intensity: New intensity value. color: [r, g, b] new light color (0-1).

Pause the physics simulation.

Start the physics simulation.

Force re-scan the asset server for available robots. Use this if new robot assets were added.

Reload a Python controller from a file on disk. Two modes, chosen automatically: - If any Action-Graph ScriptNode references this file (inputs:scriptPath), those ScriptNodes are force-recompiled so on-disk edits take effect on the running graph. This is how you iterate on a ScriptNode controller. - Otherwise the file is (re-)executed as a standalone controller. The file's directory is auto-added to sys.path. Args: file_path: Path to the Python file on disk. script_file: Alias for file_path — create_action_graph spells it that way, and both are accepted so neither spellin…

Search the NVIDIA USD asset library by text description, then load the best match. Args: text_prompt: Text description of the 3D asset to search for. target_path: Prim path for the loaded result. position: [x, y, z] world position. scale: [sx, sy, sz] scale factors.

Write a joint drive's stiffness, damping and/or max_force. Use after get_joint_config shows a drive not tracking or oscillating. Pass only the fields to change. Gains take the units get_joint_config reports (per-radian or per-meter, its `gain_units`), so a value read there can be passed back unchanged. Values must be >= 0. Errors if the joint has no drive, or its name matches several joints under prim_path. On Newton the stage is updated but the running solver keeps its old gain (`applied_to_solver: false`); tune live gains on PhysX. Args: prim_path: Robot articulation root, as passed t…

Set target joint positions on a robot via ArticulationAction. Units: radians for revolute joints, meters for prismatic joints (e.g. gripper fingers). Use get_robot_info to discover joint names, types, and limits first. After calling this, use step_simulation to advance and observe the result — do not use play_simulation + sleep. Args: prim_path: The prim path of the robot. joint_positions: List of target joint position values. joint_indices: Optional list of joint indices to set. Sets all joints if not provided.

Configure physics engine parameters. Args: gravity: Gravity vector [x, y, z]. time_step: Physics time step in seconds. gpu_enabled: Enable GPU-accelerated physics.

Advance the simulation by exactly N physics frames on a FROZEN timeline. Initialises physics on first call and operates on a paused/stopped timeline, so N is exact and observations correlate to a known frame count. Do NOT call play_simulation before or during the debug loop — step is for a frozen timeline and errors if the timeline is already playing. play is only for a final continuous run / ScriptNode demo. Args: num_steps: Number of simulation frames to step. observe_prims: Prim paths to observe. Each returns position_world, or position_local plus position_world_warning w…

Stop the physics simulation and reset to spawn state. Resets articulations and rigid bodies to their spawn pose (the state captured at first Play), like the Isaac UI Stop button. Call this to return the scene to a clean starting point before another run.

Set the transform (position, rotation, scale) of an existing object. Args: prim_path: The prim path of the object to transform. position: [x, y, z] new world position. rotation: [rx, ry, rz] new rotation in degrees. scale: [sx, sy, sz] new scale factors.