RoboClaw is an open-source embodied intelligence assistant.
🖥️ CLI |
📱 Mobile |
|---|---|
roboclaw_demo_small.mp4 |
Coming Soon |
- 2026-04-11 Web dashboard for the full embodied workflow.
- 2026-03-24 Conversational arm setup, calibration, teleoperation, data collection, training, and inference.
- 2026-03-17 Embodied framework skeleton, domain contracts, and assembly-centered onboarding controller.
- 2026-03-12 Repository created.
AI-assisted setup: ask your coding assistant:
Help me install RoboClaw from https://github.com/MINT-SJTU/RoboClaw
RoboClaw is being built in the open. We want major direction-setting choices, such as embodiment support, simulator priorities, and roadmap focus, to be discussed with the community.
You can contribute through:
Issues: bug reports, feature requests, and implementation suggestionsPull Requests: code and documentation improvements
The most useful contribution areas right now are:
- embodied AI architecture
- capability abstraction and semantic skill interfaces
- ROS2 and execution-layer integration
- simulator support and real robot adaptation
- evaluation, validation, and developer experience
If you want to contribute more actively, contact us at bozhaonanjing [[@]] gmail [[DOT]] com.
RoboClaw references and inherits part of its initial thinking from nanobot. We appreciate its lightweight practice along the OpenClaw line, which helped us build the first prototype faster and continue evolving toward embodied intelligence.
- Discord: Join the server
- WeChat official post: Coming Soon
- GitHub Issues: Create an issue
- Email: bozhaonanjing [[@]] gmail [[DOT]] com
@misc{roboclaw2026,
title = {RoboClaw: An Open-Source Embodied Intelligence Assistant},
author = {RoboClaw Contributors},
year = {2026},
howpublished = {\url{https://github.com/MINT-SJTU/RoboClaw}}
}