Sketchup MCP

Community

Connect Sketchup to Claude AI via MCP for prompt-driven 3D modeling and scene manipulation.

How to install ↓ Compare GitHub source repository ↗

Category: Dev Tools #391 of 438
Stars: ★ 487 Popular
Transport: stdio (local process)
Runtime: Python · Prebuilt binary
Credentials: No credential needed
Last commit: 5 months ago
Tools: 8

35FMRS · D

This MCP server provides powerful AI integration for Sketchup, but requires technical background and awareness of security risks.

Strongest · Setup experience 11/20 Weakest · Security and permissions 3/20

Reliability

5/20

Security and permissions

3/20

Maintenance

6/20

Documentation

10/20

Setup experience

11/20

Why each score

Reliability 5/20: This review is static and only the README is available; no source code or Server Manifest was supplied, so the MCP init handshake, tool list, and actual behavior cannot be verified. No tests, CI, or run logs provide executable evidence. The architecture description is plausible, but error handling, timeouts, and edge cases are undocumented, so a conservative score of 5 is given.
Security and permissions 3/20: The README explicitly exposes eval_ruby, which executes arbitrary Ruby code inside SketchUp, and the TCP server defaults to port 9876 with no mention of authentication, authorization, or confirmation for dangerous operations. This is a remote-command default without confirmation and falls into the 0-4 red-line band. No malware or credential theft is evidenced, hence 3.
Maintenance 6/20: The README declares an MIT license, the repository is not archived, and PyPI publishing plus contributions are mentioned. However, there is no evidence of commit history, releases, dependency updates, issue response times, or a security response channel. The status and timeliness of the 17 open issues are unknown. Star count is a discovery signal only and does not add points, so the score is 6.
Documentation 10/20: The README covers features, installation, Claude client configuration, tool list, example prompts, troubleshooting, the TCP protocol, and license. It is serviceable, but lacks detailed tool parameters, security boundaries, limitations, Python version requirements, and deployment environment specifics. This matches the 'usable but with hidden assumptions' anchor, so 10.
Setup experience 11/20: The setup path is fairly clear: install uv, install the Sketchup .rbz extension, start the extension server, and configure Claude with a JSON example. However, how to download or build the .rbz, how to run the server locally, Python version requirements, and platform compatibility are missing. There is no CI or test evidence proving the installation is repeatable, and the static calibration caps this dimension at 15; therefore 11.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessWrites / deletes local filesRuns commands or code

Best for

Users of Sketchup who want AI-assisted modeling.
Designers and developers needing automation of modeling tasks.
Users familiar with Ruby and willing to extend Sketchup functionality.

Not for

Users who do not use Sketchup.
Lightweight scenes requiring real-time rendering or complex animations.
Environments with high security requirements that disallow arbitrary code execution.

Required permissions

Create, modify, delete components within Sketchup.
Execute arbitrary Ruby code, potentially affecting the Sketchup environment.
Network communication via TCP (default port 9876).

Risks and side effects

The eval_ruby tool allows arbitrary Ruby execution, which could crash Sketchup or corrupt data.
TCP connection is unencrypted and could be intercepted.
Dependencies on both Sketchup and MCP server running simultaneously; otherwise connection fails.

Setup

Before you start

Runtime：Python · Prebuilt binary

Install uv via brew install uv.
Download or build the latest .rbz file, install the extension in Sketchup via Window > Extension Manager, and restart Sketchup.
Run the MCP server using uvx sketchup-mcp and configure Claude to use it.

claude_desktop_config.json

{
  "mcpServers": {
    "sketchup": {
      "command": "uvx",
      "args": [
        "sketchup-mcp"
      ]
    }
  }
}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json

{
  "servers": {
    "sketchup": {
      "command": "uvx",
      "args": [
        "sketchup-mcp"
      ]
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal

claude mcp add sketchup -- uvx sketchup-mcp

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

Start the Sketchup extension server via Extensions > SketchupMCP > Start Server (default port 9876), then check that tools like get_scene_info appear in your Claude client's tool list; calling get_scene_info and getting scene details confirms the connection works.

Troubleshooting

Ensure both the Sketchup extension server and the MCP server are running.
Check the Ruby Console in Sketchup for error messages.
For timeout errors, simplify requests or break them into smaller steps.

Things to try

Once connected, you can ask your AI assistant things like:

Create a simple house model with a roof and windows
Get information about the currently selected components
Make the selected component red
Move the selected component 10 units up

Tools 8

get_scene_info read-only

Gets information about the current Sketchup scene.

get_selected_components read-only

Gets information about currently selected components.

create_component writes

Create a new component with specified parameters.

delete_component destructive

Remove a component from the scene.

transform_component writes

Move, rotate, or scale a component.

set_material writes

Apply materials to components.

export_scene writes

Export the current scene to various formats.

eval_ruby writes

Execute arbitrary Ruby code in SketchUp for advanced operations.

Use cases

Create 3D models via natural language commands, e.g., a house model.

Query and modify components in the current scene.

Apply materials or change component properties.

Export scenes to various formats.

Supported clients

Claude Desktop

Listed from the project's documentation, not tested by this site.

Overview

SketchupMCP connects Sketchup to Claude AI through the Model Context Protocol (MCP), allowing Claude to directly interact with and control Sketchup. This integration enables prompt-assisted 3D modeling, scene creation, and manipulation. The system consists of two components: a Sketchup extension that creates a TCP server within Sketchup, and a Python MCP server that connects to it. It offers two-way communication, component manipulation, material control, scene inspection, selection handling, and Ruby code evaluation.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 62.9k · Tools 2 Compare with this →

NPM Sentinel MCP Server 79 · B

AI-powered NPM package analysis MCP server

★ 18 · Tools 19 Compare with this →

MartinLoop MCP Server 79 · B

An independent control layer that gives AI coding agents budgets, brakes, and receipts

★ 410 Compare with this →

GitHub MCP Server 76 · B

GitHub's official MCP server for managing repos, issues, PRs, and workflows via natural language

★ 33.5k · Tools 15 Compare with this →

Compare with other MCP servers

Source revision aa096f04d3d7 Data synced 2026-10-12 Read the FMRS scoring method