Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.cyberwave.com/llms.txt

Use this file to discover all available pages before exploring further.

A digital twin is Cyberwave’s unit of “physical thing made programmable.” It carries everything the platform needs to know about a robot, a sensor, a drone — or even a passive object like a table or a shelf — and keeps it in lockstep with the real world. One twin gives you a 3D model, physical properties and capabilities, sensor streams, a live link to hardware, and a stable identity you can reference from code, MQTT, or the dashboard. 
from cyberwave import Cyberwave

cw = Cyberwave()
robot = cw.twin("unitree/go2")
That’s it — the twin is alive in your environment, ready to be simulated, controlled, or paired with the real robot.

​
Why a digital twin?

One object, one source of truth

Geometry, kinematics, sensors, telemetry, and access control — all attached to a single entity.

Sim-to-real, real-to-sim

The same twin runs in simulation and against physical hardware. Swap modes with one call: cw.affect("simulation") or cw.affect("live").

Hardware in minutes, not weeks

Pair any device with cyberwave pair — Cyberwave installs the matching driver and starts streaming.

Capability-driven UX

UI panels, SDK classes, and controllers light up automatically based on what the twin can do.

​
What’s inside a twin

A twin’s capabilities are computed from its underlying asset and unlock features across the platform.
AspectWhat it coversLearn more
Physical propertiesSize, mass, payload, power source and capacity, geometry, 3D meshCapabilities map
LocomotionWheeled, tracked, legged, aerial, surface; max linear / angular velocityLocomotion config
ManipulationJoints, end-effectors, grippers, payload capacityActuation features
SensorsRGB, depth, LiDAR (2D / 3D / 4D), microphones, maps — with offsets and FOVSensor capabilities
NavigationAutonomy level, obstacle avoidance, waypoints and missionsNavigation capabilities
IdentityA unique slug like acme/twins/arm-station-1Slug system
RecordingPer-twin cloud recording for video, audio, telemetryCloud recording
Access controlPer-twin and per-workspace roles and visibilityAccess control

​
Use a twin from the SDK

Every twin is reachable from the Python SDK with a one-liner. The SDK returns the most specific twin class based on its capabilities — FlyingTwin for drones, GripperTwin for arms, CameraTwin for cameras, and so on. 
from cyberwave import Cyberwave

cw = Cyberwave()

drone = cw.twin("dji/dji-mini-4-pro")
drone.takeoff()

arm = cw.twin("the-robot-studio/so101")
arm.joints.set("shoulder_pan", 45, degrees=True)

dog = cw.twin("unitree/go2")
dog.move_forward(distance=1.0)
You can also fetch existing twins by their unified slug: 
arm = cw.twin(twin_id="acme/twins/arm-station-1")
Twins also speak MQTT and REST if you prefer raw protocols, or want to drive them from C++ or another language.

​
A growing catalog of hundreds of twins

The Cyberwave catalog ships with hundreds of pre-built twins — robot arms, quadrupeds, drones, AMRs, cameras, microphones, LiDARs — and grows every week. Each catalog entry comes with a URDF, a 3D mesh, sensor definitions, and the right driver, so you go from “I have hardware” to “it’s streaming” without writing integration code.

Browse the catalog

Filter by category, vendor, or capability and preview each twin in 3D.

Catalog feature reference

How catalog assets, slugs, and twin creation fit together.

​
Add your own twin

Bring a new robot or sensor in minutes. The easiest path is to let an agent do the heavy lifting:

Claude Skill

Run /cyberwave in Claude Code and describe your hardware — the skill scaffolds the URDF, asset metadata, and driver for you.

MCP Server

Use Cyberwave’s MCP server from Cursor, Claude Desktop, or any MCP client to create twins, upload assets, and wire up drivers conversationally.
You can also upload a URDF ZIP manually — see the catalog upload guide.

Hello Robot

Spin up your first twin in the browser in 5 minutes.

Connect real hardware

Pair a physical robot or sensor with its twin via the edge stack.

Digital Twins reference

Full capability map, sensor schema, and feature-to-capability matrix.