> ## 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.

# Cyberwave Claude Skill

> Build Physical AI applications with Claude Code using the Cyberwave skill.

## Overview (Stub)

The Cyberwave Claude skill guides you through building a Physical AI application — controlling robots, reading sensors, managing digital twins, and streaming video.

When invoked, it walks you through three paths:

* **Python SDK** (recommended) — installation, authentication, twin connection, joint control, camera capture, video streaming, workflows, and alerts
* **C++ SDK** — coming soon; redirects to Python SDK or raw APIs in the meantime
* **APIs directly** — REST (HTTPS, Bearer token) and MQTT (real-time joint control, telemetry, WebRTC signalling) for any language

## Installation

**Global** (available in all projects):

```bash theme={null}
git clone https://github.com/cyberwave-os/cyberwave-skill ~/.claude/skills/cyberwave
```

**Project-local:**

```bash theme={null}
git clone https://github.com/cyberwave-os/cyberwave-skill .claude/skills/cyberwave
```

## Usage

In any Claude Code session, run:

```
/cyberwave
```

Describe what you want to build and Claude will guide you through environment setup, interface selection, and implementation.

## Requirements

* [Claude Code](https://claude.ai/claude-code)
* A Cyberwave account — [sign up](https://cyberwave.com)
* Python 3.11+ and FFMPEG (for the Python SDK path)