ROS机器人MCP工具

Requirements

- Node.js ≥ 20 (the only hard requirement — npx agenticros installs pnpm itself if missing) - ROS 2 Humble or Jazzy if you plan to use the real-robot stack or any simulation (the CLI sources /opt/ros/<distro>/setup.bash and runs colcon build for you) - OpenClaw gateway only if you also want the OpenClaw web UI / chat / teleop adapter

Install

You only need one command. The agenticros CLI handles everything else — installing the ROS 2 workspace, building the MCP server, registering the OpenClaw plugin, and wiring up your robot config.

npx agenticros

That's it. Run it on any machine with Node ≥ 20, no git clone required. The first run launches the interactive menu:

╔──────────────────────────────────────────────────╗
║  AgenticROS - agentic AI for ROS-powered robots  ║
╚──────────────────────────────────────────────────╝

? What would you like to do?
  Launch with real robot
❯ Launch with simulation
  First-time setup (workspace + OpenClaw plugin + API key)
  Stop everything
  Doctor (health check)
  Configure (API keys, namespace, transport)
  Tail logs

Pick First-time setup once (workspace + OpenClaw plugin + API key, all idempotent), then choose how you want to run:

Once a stack is up, point any of the supported agents — OpenClaw, Claude Code, Claude Desktop / Dispatch, or Gemini CLI — at the same robot and start talking to it. The CLI tracks what it spawned (pidfiles + logs under /tmp/agenticros-*), so Stop everything cleanly tears the demo down.

Prefer scripted invocations? Every menu item maps to a direct command:

npx agenticros init             # one-time workspace + plugin + API key
agenticros up real              # real robot stack
agenticros up sim-amr           # simulated AMR (Gazebo + RViz, headless on Jetson)
agenticros up sim-arm           # simulated 6-DOF arm
agenticros mode <real|sim>      # swap the active config profile (namespace, transport)
agenticros doctor               # coloured health check
agenticros down                 # stop everything we started

Full CLI reference: packages/agenticros-cli/README.md.

Contributing / building from source

Hacking on the packages themselves? Clone and use the local checkout — the CLI auto-detects the workspace and uses live sources instead of the bundled snapshot:

git clone https://github.com/PlaiPin/agenticros && cd agenticros
pnpm install && pnpm build
./agenticros                    # repo-local CLI shim, same menu as `npx agenticros`

For the OpenClaw plugin specifically, point the gateway at this repo's packages/agenticros and configure under plugins.entries.agenticros.config. Recommended: OpenClaw 2026.3.11+ — routes work at http://127.0.0.1:18789/plugins/agenticros/ (config, teleop). For local dev without token auth: node scripts/setup-openclaw-local.cjs then restart the gateway. Older gateways needing token auth: run node scripts/agenticros-proxy.cjs 18790 and open http://127.0.0.1:18790/plugins/agenticros/. See docs/openclaw-releases-and-plugin-routes.md and docs/teleop.md.

See docs/ for robot setup, skills, teleop, simulation internals, and Docker.

1. Build + pack the plugin so it works in the sandbox's offline-npm env

pnpm install && pnpm build pnpm --filter @agenticros/agenticros deploy --prod /tmp/agenticros-deploy rm -f /tmp/agenticros-deploy/node_modules/.pnpm/node_modules/@agenticros/agenticros

3. Register + configure the plugin inside the sandbox (HOME=/sandbox is required)

docker exec -u sandbox -e HOME=/sandbox "$CONTAINER" \ openclaw plugins install -l /sandbox/agenticros # Ctrl-C once it starts logging "ROS2 transport status:"

Memory (optional)

AgenticROS can give every adapter a shared, persistent, cross-process long-term memory so facts you teach the robot from one agent are immediately available in the others — Claude Desktop, Claude Code, Gemini CLI, OpenClaw chat. Off by default. Two backends:

• local — zero deps, JSON-on-disk at ~/.agenticros/memory.json, keyword + recency search. Enable with one config flag.
• mem0 — semantic search via the pure-Node mem0ai package (pnpm add mem0ai); file-backed vector store at ~/.mem0/vector_store.db (shared across all processes on the host, no server to run); embedder auto-detects Ollama (http://localhost:11434) → OPENAI_API_KEY → clear error.

When enabled, every adapter exposes four tools — memory_remember, memory_recall, memory_forget, memory_status. OpenClaw chats additionally get a system-context section that lists recently-remembered facts and instructs the LLM to call memory_recall before answering personal-context questions like "what do I have for X?" or "what's my Y?".

Memory is namespaced by robot.namespace so adapters talking to the same robot share the same store. See docs/memory.md for ready-to-paste recipes, cross-process verification steps, and troubleshooting. OpenClaw users can also enable it from the web config UI at /agenticros/config (Memory section).

Quickest path to try it (fully local, no API keys):

pnpm add mem0ai
ollama pull nomic-embed-text   # ~270 MB embedder model

Add { "memory": { "enabled": true, "backend": "mem0" } } to ~/.agenticros/config.json (or the OpenClaw config UI). Restart the gateway / MCP client. Then ask Claude Desktop "remember that I have a RealSense D435i for eyes" and ask OpenClaw "what do I have for eyes?" — same fact, both agents.

Claude Code CLI (terminal)

1. Register MCP (project scope, from repo root):

   claude mcp add --transport stdio --scope project agenticros -- node packages/agenticros-claude-code/dist/index.js