A comprehensive toolkit for initializing, configuring, and previewing MIPI CSI, USB, and depth cameras on the RDK X5 hardware platform.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install rdk-x5-camera
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install rdk-x5-camera using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The rdk-x5-camera skill is a specialized utility designed for the RDK X5 platform to manage a diverse range of imaging hardware. It enables developers to seamlessly interface with MIPI CSI sensors like the IMX219 or OV5647, standard USB UVC cameras, and advanced depth cameras including Intel RealSense and ZED. By utilizing this within the ecosystem of Openclaw Skills, users can bridge raw sensor data to high-level applications through a unified ROS2-based workflow.
Beyond simple initialization, this skill provides integrated tools for image encoding and WebSocket streaming. This allows for low-latency, real-time video previews directly in a web browser, making it an essential component for robotics and vision-based prototyping. It streamlines the complex task of hardware-level camera orchestration into a set of repeatable, automated commands.
To begin using this camera utility within Openclaw Skills, ensure your hardware is connected and run the following commands:
# Initialize environment
source /opt/tros/humble/setup.bash
# For MIPI CSI cameras
ros2 launch mipi_cam mipi_cam.launch.py
# For USB Cameras (MJPEG mode)
ros2 launch hobot_usb_cam hobot_usb_cam.launch.py usb_video_device:=/dev/video0
# Start Web Preview
ros2 launch hobot_codec hobot_codec_encode.launch.py &
ros2 launch websocket websocket.launch.py websocket_image_topic:=/image_jpeg websocket_only_show_image:=true
The skill manages data flow through standard ROS2 topics and localized system nodes:
| Data Type | Topic / Path | Description |
|---|---|---|
| Raw Image | /image_raw |
Unprocessed stream from the primary camera sensor |
| Compressed Image | /image_jpeg |
JPEG encoded stream for web distribution |
| Device Nodes | /dev/video* |
Linux system nodes for USB and MIPI devices |
| Web Interface | http://<RDK_IP>:8000 |
Local endpoint for real-time visualization |
| ISP Params | /app/multimedia_samples/tuning_tool |
Directory for image signal processor tuning configs |
Loading
A powerful skill for generating, animating, and editing high-quality AI videos using Kling models via the Atlas Cloud API.

A sophisticated AI companion skill that leverages Adlerian psychology to transform negative emotions into actionable micro-steps.

A specialized tool for retrieving and formatting A-share financial performance express reports into clean Markdown tables.

A local skill management system that enables AI agents to automatically route tasks based on defined trigger keywords.

A professional monitoring tool that tracks website availability and generates daily visual performance reports.

An AI agent skill for designing hyper-realistic, terrain-aware Colombian travel itineraries based on verified entry rules and regional playbooks.








































