机器人MCP服务器

一个用于统一控制物理机器人、Gazebo 仿真环境和 Unity/VRChat 虚拟栈的 MCP 服务器,支持 FastMCP、stdio/HTTP 接口及自动化流水线。

  • by sandraschi
  • Python
  • 本地部署
  • 12 星标
  • 更新于
在 GitHub 查看 sandraschi/robotics-mcp

使用说明

Robotics MCP Server

📖 Installation Guide — quick start, manual setup, and troubleshooting

Author: FlowEngineer sandraschi

One MCP server for physical robots (Dreame vacuums, Yahboom ROS, Elegoo, Nori A3), Gazebo, and virtual stacks (Unity, VRChat). FastMCP, stdio + HTTP. Optional watchfiles restarts for dev or unattended runs (not an SLA).

Doc Contents
Technical README Prerequisites, hardware, install, packaging, Cursor, config, composed MCPs, testing, troubleshooting
Usage and API Tool examples, web UI, HTTP endpoints
Android Doctrine Safety, legal, and operational standards for the android fleet
Setup Prerequisites Unity, VRChat, peer MCP servers when you need the full stack
Crash protection (watchfiles) Auto-restart and health checks
Autonomous Robotics Pipeline 11-stage CAD -> OpenUSD -> Isaac Lab (RSL-RL / GR00T N1.5 / Newton) -> Genie 4 World Model Dojo -> ROS 2 -> Active Inference (saschibot & Nori A3)
Gemini Robotics Distillations 18-part synthesis of CAD/USD pipelines, Isaac Lab RL, Genie 4 World Model Dojo, ros2_control, and Active Inference (FEP)
Hardware Reference Subsystem histology (TELESCOPIC_LIFT, ROLLING_BASE, ARMS, GRIPPERS_HANDS, DOCK_STATION, HEAD_SENSORS_SCREEN)

Status: Industrial v1.4.1 (2026-04-14) | SOTA v1.4.1 Compliance | Benny Protocol Secured.

Quick Start

git clone https://github.com/sandraschi/robotics-mcp
cd robotics-mcp
just

This opens an interactive dashboard showing all available commands. Run just bootstrap to install dependencies, then just serve or just dev to start.

Manual Setup

If you don't have just installed:

Install

uvx robotics-mcp
pip install robotics-mcp

Development clone: see Installation in the technical README.

Contributing

See CONTRIBUTING.md.

🛡️ Industrial Quality Stack

This project adheres to SOTA 14.1 industrial standards for high-fidelity agentic orchestration:

  • Python (Core): Ruff for linting and formatting. Zero-tolerance for print statements in core handlers (T201).
  • Webapp (UI): Biome for sub-millisecond linting. Strict noConsoleLog enforcement.
  • Protocol Compliance: Hardened stdout/stderr isolation to ensure crash-resistant JSON-RPC communication.
  • Automation: Justfile recipes for all fleet operations (just lint, just fix, just dev).
  • Security: Automated audits via bandit and safety.

License

MIT see LICENSE.

Acknowledgments

FastMCP; ROS community; Unity3D, VRChat, World Labs Marble/Chisel; MCP ecosystem contributors.