AgenticROS: Connects ROS with OpenClaw, Claude (code, desktop, dispatch), and Google Gemini

[Chris_Matthieu]

I’ve been working on a new open source project called AgenticROS. It adds ROS2 support to multiple AI agents: OpenClaw, NemoClaw, Claude Code, Claude desktop & Dispatch, Google Gemini, and others.

AgenticROS supports four deployment modes depending on where your AI agent runs relative to the robot.

Mode A - Same Machine

Your AI agent runs on the robot’s computer. Plugin talks to ROS2 via local DDS — no network transport. Best for: Single-robot setups, embedded deployments, edge devices.

Mode B - Local Network

Your AI agent (e.g. OpenClaw, NemoClaw, Google Gemini, Claude Code, or Claude desktop / Dispatch) on a separate machine (laptop, server). Adapter connects to rosbridge_server on the robot via WebSocket over LAN. Best for: Development, testing, multi-robot labs.

Mode C - Cloud / Remote

AI agent on a cloud server; robot behind NAT. WebRTC with STUN/TURN for NAT traversal. AgenticROS Agent Node on robot. Best for: Production, remote operations, fleet management.

Mode D - Zenoh

Robot uses ROS 2 with Zenoh RMW. Plugin connects via zenoh-ts (WebSocket to Zenoh router). No rosbridge required. Best for: Robots already on Zenoh, low-latency pub/sub.

SKILLS

AgenticROS also has an open plugin Skills architecture. The AgenticROS plugin loads optional skill packages at startup. Each skill registers its own tools (e.g. follow_robot, follow_me_see) and reads config from config.skills.<skillId>. You install skills by adding them to skillPackages (npm names) or skillPaths (local directories) in the plugin config, then restarting the gateway.

A curated list of AgenticROS skills helps robotics developers discover available skills and add new ones via pull request. The Follow Me skill (agenticros-skill-followme) is the reference implementation: depth-based person following with optional Ollama/VLM, turn-to-follow and search-when-lost. Use it as a template to build and share your own skills.

Try It

Send a message to your robot (via OpenClaw, NemoClaw, Claude Code, Claude Desktop / Dispatch, Google Gemini, or any supported agent):

“Move forward 1 meter” — publishes velocity to /cmd_vel

“Follow Me” — skill to follow a person (depth + optional VLM)

“Navigate to the kitchen” — sends a Nav2 goal

“What do you see?” — captures a camera frame

“Check the battery” — reads /battery_state/estop — emergency stop (bypasses AI)

DEMO VIDEO

AgenticROS (ROS + OpenClaw) Demo

[nathan]

This is amazing Chris. Can we connect? I am working on something similar for my research.

[devx]

Interested in this as well, will give it a try

[Chris_Matthieu]

Yes, love to connect!

chris.matthieu@realsenseai.com

[Chris_Matthieu]

Awesome! Please send feedback.

[Chris_Matthieu]

AgenticROS now officially supports Nvidia’s NemoClaw (in addition to OpenClaw, Claude, and Gemini) - #ros | AgenticROS

[Chris_Matthieu]

New FindObject Physical AI skill added to AgenticROS!

Your ROS robot can now look for physical items in your space using Nvidia’s NemoClaw or OpenClaw, Claude, or Google Gemini.

Try it: https://github.com/agenticros/agenticros-skill-find

[Chris_Matthieu]

AgenticROS now has a persistent memory service that runs across AI Agents!

When the AI agent memory service is enabled, every adapter — OpenClaw, Claude Code, Claude Desktop / Dispatch, and Gemini — exposes the same four tools backed by a shared, file-backed store. A fact remembered from one adapter is immediately recallable from any other on the same host.

Tool	Description
memory_remember	Store a fact (content, tags, namespace) in cross-adapter long-term memory
memory_recall	Search memories by free-text query — returns ranked matches across adapters
memory_forget	Delete a memory by id, query, or namespace
memory_status	Health check: enabled, backend, namespace, record count, last write timestamp

AgenticROS — Agentic Robotics

[Chris_Matthieu]

[Chris_Matthieu]

[Fedor]

Cool stuff! Can it operate at real life like, was it tested?

[Chris_Matthieu]

NPX AGENTICROS

With a single command, AgenticROS can install itself on a ROS-based system and launch an interactive setup wizard. No manual repo cloning. No stitching together setup scripts by hand. The CLI walks you through first-time setup, robot configuration, OpenClaw integration, skills, simulation, logs, health checks, and clean shutdowns.

In addition to AI Agentis controlling real ROS-powered robots, AgenticROS now supports Rviz and Gazebo simulations controlled by OpenClaw, Claude, and Gemini!

More info: AgenticROS Just Got Much Easier to Install, Run, and Demo