MCP server that connects Claude and other MCP clients to ROS and ROS 2 robots through rosbridge for topic, service, action, parameter, sensor, and robot state workflows.
ROS MCP Server can expose robot observation and control surfaces to an AI client, including topic publishing, topic subscription, service calls, action calls, parameter updates, sensor reads, and robot state inspection., Treat every topic, service, action, and parameter exposed through rosbridge as potentially capable of moving hardware, changing robot behavior, altering safety limits, or interrupting active tasks., Validate workflows in simulation or with the robot immobilized before allowing commands on mobile bases, manipulators, drones, industrial equipment, or other physical systems., Keep an operator, emergency stop, physical safety perimeter, and robot-specific lockout procedures in place for live hardware., Restrict rosbridge to trusted networks, approved clients, and the minimum reachable robot namespace needed for the task., Review generated commands before execution when using custom message types, services, actions, or parameters that the model has inferred from live ROS metadata.
Privacy notes
ROS topics and services can expose camera frames, sensor data, maps, robot poses, facility layouts, task names, device identifiers, hostnames, IP addresses, and operational telemetry., Images, logs, maps, and robot state returned to the MCP client may also be sent to the model provider or stored by the MCP client, terminal, or chat application., rosbridge traffic crosses the local network or VPN; protect robot networks from untrusted clients and avoid exposing rosbridge directly to the public internet., Review launch files, client configs, environment variables, and connection prompts for credentials, robot addresses, VPN details, or internal network names before sharing transcripts.
Author
robotmcp
Submitted by
oktofeesh1
Claim status
unclaimed
Last verified
2026-06-06
Decision playbook
Review trust signals before you adopt
Signals are present but mixed. Use the checklist below to confirm the source and operational safety for your environment.
Compare context
Selected
0
Current score
63
Baseline
—
Delta
No baseline selected
No major trust-signal divergence detected in the current selection.
Source and provenance checks
Needs review
Confirm ownership and provenance before trusting install instructions.
Source link availableRequired
Open the canonical repository and verify ownership.
Done
Source provenance statusRequired
Marked as source-backed.
Done
Metadata reviewed
No reviewed flag detected in metadata.
Pending
Safety and privacy checks
Complete
Validate risk disclosures before installation or API wiring.
Safety notes presentRequired
Review the listed safety guidance before running commands.
Done
Privacy notes presentRequired
Review data handling notes before connecting accounts or secrets.
Done
Trust level risk gateRequired
Trust level does not block evaluation.
Done
Package and install checks
Needs review
Check package metadata and artifact integrity signals.
Install payload available
Install or copy payload is available for review.
Done
Package verification flag
No package verification flag provided.
Pending
Checksum metadata
No checksum provided for downloaded artifact.
Pending
Compare-driven decision checks
Needs review
Use compare context to validate trade-offs before adoption.
Compare tray has multiple entries
Add at least one more entry to compare trust differences.
6 safety and 4 privacy notes across 6 risk areas. Review closely: credentials & tokens, network access, third-party handling.
6 areas
SafetyExecution & processesROS MCP Server can expose robot observation and control surfaces to an AI client, including topic publishing, topic subscription, service calls, action calls, parameter updates, sensor reads, and robot state inspection.
SafetyGeneralTreat every topic, service, action, and parameter exposed through rosbridge as potentially capable of moving hardware, changing robot behavior, altering safety limits, or interrupting active tasks.
SafetyExecution & processesValidate workflows in simulation or with the robot immobilized before allowing commands on mobile bases, manipulators, drones, industrial equipment, or other physical systems.
SafetyGeneralKeep an operator, emergency stop, physical safety perimeter, and robot-specific lockout procedures in place for live hardware.
SafetyNetwork accessRestrict rosbridge to trusted networks, approved clients, and the minimum reachable robot namespace needed for the task.
SafetyExecution & processesReview generated commands before execution when using custom message types, services, actions, or parameters that the model has inferred from live ROS metadata.
PrivacyTelemetryROS topics and services can expose camera frames, sensor data, maps, robot poses, facility layouts, task names, device identifiers, hostnames, IP addresses, and operational telemetry.
PrivacyThird-party handlingImages, logs, maps, and robot state returned to the MCP client may also be sent to the model provider or stored by the MCP client, terminal, or chat application.
PrivacyNetwork accessrosbridge traffic crosses the local network or VPN; protect robot networks from untrusted clients and avoid exposing rosbridge directly to the public internet.
PrivacyCredentials & tokensReview launch files, client configs, environment variables, and connection prompts for credentials, robot addresses, VPN details, or internal network names before sharing transcripts.
Disclosure: Open-source Apache-2.0 MCP server published as the `ros-mcp` PyPI package and designed for ROS and ROS 2 robots through rosbridge.
Safety notes
ROS MCP Server can expose robot observation and control surfaces to an AI client, including topic publishing, topic subscription, service calls, action calls, parameter updates, sensor reads, and robot state inspection.
Treat every topic, service, action, and parameter exposed through rosbridge as potentially capable of moving hardware, changing robot behavior, altering safety limits, or interrupting active tasks.
Validate workflows in simulation or with the robot immobilized before allowing commands on mobile bases, manipulators, drones, industrial equipment, or other physical systems.
Keep an operator, emergency stop, physical safety perimeter, and robot-specific lockout procedures in place for live hardware.
Restrict rosbridge to trusted networks, approved clients, and the minimum reachable robot namespace needed for the task.
Review generated commands before execution when using custom message types, services, actions, or parameters that the model has inferred from live ROS metadata.
Privacy notes
ROS topics and services can expose camera frames, sensor data, maps, robot poses, facility layouts, task names, device identifiers, hostnames, IP addresses, and operational telemetry.
Images, logs, maps, and robot state returned to the MCP client may also be sent to the model provider or stored by the MCP client, terminal, or chat application.
rosbridge traffic crosses the local network or VPN; protect robot networks from untrusted clients and avoid exposing rosbridge directly to the public internet.
Review launch files, client configs, environment variables, and connection prompts for credentials, robot addresses, VPN details, or internal network names before sharing transcripts.
Prerequisites
Python 3.10 or newer on the machine running the MCP client.
A ROS or ROS 2 robot, simulator, or development machine with rosbridge available.
Network access from the MCP server machine to the rosbridge WebSocket endpoint.
Familiarity with the robot's topics, services, actions, parameters, frames, and safety limits.
A simulation, test bench, or supervised robot environment for validating prompts before live operation.
Schema details
Install type
cli
Troubleshooting
No
Source repository stats
Scope
Source repo
Collection metadata
Estimated setup
30 minutes
Difficulty
advanced
Tool listing metadata
Disclosure
Open-source Apache-2.0 MCP server published as the `ros-mcp` PyPI package and designed for ROS and ROS 2 robots through rosbridge.
ROS MCP Server connects MCP clients to robots and simulations running ROS or
ROS 2. It runs on the AI client machine, connects to rosbridge on the robot or
ROS host, and lets Claude inspect and use live ROS topics, services, actions,
parameters, sensor streams, and robot state.
Use it when Claude needs to explore a ROS graph, debug a robot, operate a
simulator, or help with supervised robotics workflows without modifying the
robot's existing source code. It is best suited to robotics teams that already
understand their robot's ROS interfaces and safety limits.
These sources were reviewed on 2026-06-06. Prefer the live repository,
installation guide, PyPI metadata, license, README, package manifest, MCP
registry metadata, rosbridge setup guide, server entrypoint, and topic tool
implementation for current setup and behavior details.
Features
Connect an MCP client to a ROS or ROS 2 system through rosbridge.
Discover available topics, services, actions, parameters, and message types.
Publish and subscribe to ROS topics.
Call ROS services and actions.
Inspect and set parameters.
Read sensor data and monitor robot state.
Work with custom ROS message and service types exposed through the robot's
rosbridge environment.
Run from Claude Code, Claude Desktop, Codex CLI, Gemini CLI, ChatGPT,
Cursor, and other MCP-compatible clients.
Installation
For MCP clients that can run Python packages with uvx, use:
Then connect from the MCP client to the robot host or simulator running
rosbridge.
Use Cases
Ask Claude to inspect topics and services on a ROS 2 robot.
Debug robot state by reading sensors, parameters, and live ROS metadata.
Operate simulated robots from natural-language prompts.
Explore custom ROS message and service types before writing integration code.
Test robot services or actions under human supervision.
Bridge AI assistant workflows into ROS labs, demos, and test benches.
Safety and Privacy
This server can connect Claude to physical robot control surfaces. Test in
simulation first, narrow the available ROS namespace, and keep human supervision
and emergency-stop procedures in place before using it with live hardware.
Rosbridge should be reachable only from trusted hosts on a protected local
network or VPN. Avoid exposing it directly to the public internet, and avoid
connecting an AI client to production, industrial, mobile, aerial, or heavy
robotics systems without explicit operational approval.
ROS data can include camera frames, maps, facility layout, robot location,
hardware identifiers, network details, telemetry, and logs. Treat transcripts
and MCP responses as sensitive when they include robot observations or commands.
Duplicate Check
Existing MCP entries cover developer tools, browsers, databases, observability,
and automation servers, but no ROS MCP Server, robotmcp/ros-mcp-server,
ros-mcp PyPI package, or matching rosbridge robotics MCP entry was found in
content/mcp.
MCP server that connects Claude and other MCP clients to ROS and ROS 2 robots through rosbridge for topic, service, action, parameter, sensor, and robot state workflows.
MCP server from HttpRunner that exposes UIXT device and browser automation tools for Android, iOS, Harmony, and Web testing through the `hrp mcp-server` command.
MCP server for running AppleScript and JavaScript for Automation on macOS, with a searchable knowledge base of automation recipes for apps, files, browsers, system settings, Terminal, and productivity workflows.
✓ROS MCP Server can expose robot observation and control surfaces to an AI client, including topic publishing, topic subscription, service calls, action calls, parameter updates, sensor reads, and robot state inspection.
Treat every topic, service, action, and parameter exposed through rosbridge as potentially capable of moving hardware, changing robot behavior, altering safety limits, or interrupting active tasks.
Validate workflows in simulation or with the robot immobilized before allowing commands on mobile bases, manipulators, drones, industrial equipment, or other physical systems.
Keep an operator, emergency stop, physical safety perimeter, and robot-specific lockout procedures in place for live hardware.
Restrict rosbridge to trusted networks, approved clients, and the minimum reachable robot namespace needed for the task.
Review generated commands before execution when using custom message types, services, actions, or parameters that the model has inferred from live ROS metadata.
✓AI Game Developer can create, move, copy, modify, and delete Unity assets, scenes, GameObjects, components, scripts, packages, and generated project files.
Tools include dynamic C# script execution, C# reflection method lookup and calls, package installation/removal, Unity test execution, editor state changes, play mode control, screenshots, and profiler access.
The MCP server supports streamable HTTP and stdio; HTTP deployments should require a bearer token and stay bound to trusted interfaces.
Server variables include optional authorization and webhook settings; webhook endpoints can receive tool, prompt, resource, connection, and authorization events.
Unity runtime connections can expose compiled game state or in-game behavior to an MCP client, not just editor-only project data.
Use source control, backups, tool filtering, and explicit review before allowing destructive tools such as asset deletion, package removal, script execution, or reflection calls.
✓HttpRunner UIXT MCP Server can control real devices and browsers through taps, swipes, text input, hardware buttons, selectors, OCR, CV, and AI action tools.
App tools can launch, terminate, cold launch, install, uninstall, clear data, and inspect foreground apps on connected devices.
Screen tools can capture screenshots, screen recordings, screen size, and UI hierarchy/source data.
Media and utility tools can modify test-device albums, close popups, wait randomly, and interact with system UI state.
Simulated gesture options and anti-risk-style behavior should be used only in authorized test automation environments.
Require human review before running actions against logged-in accounts, payment flows, device settings, production apps, or customer data.
✓macOS Automator MCP executes AppleScript or JavaScript for Automation through `osascript`, either from inline script content, an absolute script path, or a knowledge-base script ID.
AppleScript and JXA can control local applications, read and write files, run shell commands through scripting additions, manipulate browser tabs, send keystrokes, change settings, and trigger app-specific actions depending on macOS permissions.
Do not grant broad Automation or Accessibility permissions to the host process unless you trust the MCP client, prompts, and scripts it will run.
Review generated scripts before execution, use short timeouts for untrusted automations, and avoid running scripts against Mail, Messages, Contacts, Calendar, browsers, password managers, terminals, or cloud-sync folders without explicit approval.
Knowledge-base scripts and custom local knowledge-base overrides can change the behavior of reusable script IDs; review local overrides before relying on them.
Use separate macOS user accounts or test machines when experimenting with destructive file, shell, browser, or system-setting automation.
Privacy notes
✓ROS topics and services can expose camera frames, sensor data, maps, robot poses, facility layouts, task names, device identifiers, hostnames, IP addresses, and operational telemetry.
Images, logs, maps, and robot state returned to the MCP client may also be sent to the model provider or stored by the MCP client, terminal, or chat application.
rosbridge traffic crosses the local network or VPN; protect robot networks from untrusted clients and avoid exposing rosbridge directly to the public internet.
Review launch files, client configs, environment variables, and connection prompts for credentials, robot addresses, VPN details, or internal network names before sharing transcripts.
✓Tool calls may expose Unity project paths, asset names, scene hierarchy, serialized object data, scripts, logs, screenshots, profiler metrics, test output, package metadata, and runtime state.
MCP config files and Unity plugin config can contain server URLs, connection modes, bearer tokens, authorization settings, enabled tool IDs, and cloud or local endpoint details.
Optional webhooks can receive tool, prompt, resource, connection, authorization, and token-bearing request data.
Screenshots from Game View, Scene View, cameras, or isolated GameObjects may include proprietary artwork, level design, UI, debug overlays, or unreleased game content.
Generated skills and AI-client configuration files can reveal available tools, project structure, installed packages, and local workflow assumptions.
✓Screenshots, screen recordings, UI hierarchy, package names, foreground app names, OCR text, selector data, and logs can expose sensitive app or account state.
Connected device identifiers, serial numbers, app package names, browser pages, and test artifacts may be visible to the MCP client.
Input and paste actions can expose credentials, one-time codes, chat text, forms, or clipboard-like data if used on personal sessions.
Test reports, MCP responses, and debugging logs may retain captured UI state after the session ends.
Keep test devices resettable, avoid personal accounts, and isolate artifacts produced by automated UI runs.
✓Scripts can expose local file paths, file contents, clipboard data, browser URLs and page data, email subjects, contacts, calendar events, messages, app state, system settings, shell output, and account-specific metadata.
Error responses and optional debugging settings can include attempted script content, script paths, substitution logs, stderr, and execution details.
The local knowledge base defaults to `~/.macos-automator/knowledge_base` when configured and may contain private automation recipes or paths.
MCP clients, model providers, terminal history, and application logs may retain script inputs and outputs, including data read from local apps.
Prerequisites
Python 3.10 or newer on the machine running the MCP client.
A ROS or ROS 2 robot, simulator, or development machine with rosbridge available.
Network access from the MCP server machine to the rosbridge WebSocket endpoint.
Familiarity with the robot's topics, services, actions, parameters, frames, and safety limits.
Unity project using Unity 2022.3 or newer for the current package metadata.
Unity Hub or a local Unity Editor installation.
Node.js 20.19 or newer, or Node.js 22.12 or newer, for `unity-mcp-cli`.
An MCP client such as Claude Code, Claude Desktop, Codex, Cursor, Gemini CLI, GitHub Copilot, Cline, or another supported client.
Go 1.23 or newer for building the current v5 CLI from source.
A test Android, iOS, Harmony, or browser automation environment configured for HttpRunner UIXT.
A trusted MCP client that can launch local stdio servers.
Review of ADB, go-ios, WebDriver, OCR, computer-vision, and AI-action dependencies required by your target platform.
macOS on the machine running the MCP server.
Node.js 24 or newer for the published npm package.
Automation permissions for the terminal, editor, or app that launches the MCP server.
Accessibility permissions when scripts need UI scripting, clicks, keystrokes, or System Events access.