Suits web developers, architecture hobbyists and agent builders; you should be comfortable with Node.js, and with React and Three.js if you plan to use the packages.

What can you build with editor?

Sketch a floor plan in 3D and keep the project saved locally on your machine. Connect Claude Code or Codex to Pascal and ask the agent to build or modify a building scene through MCP. Check whether a piece of furniture fits a room's footprint using the furniture-fit skill. Embed the 3D viewer in your own React or Next.js app to display building models. Study a working example of a React Three Fiber editor built on Zustand and an MCP server.

How do I get started with editor?

With Node.js 22.13 or newer, run the npx command to start the editor locally. To connect an agent, install the CLI globally and point your agent at `pascal mcp connect`, or install the Claude Code plugin. Commands from the README: npx @pascal-app/cli editor; npm install --global @pascal-app/cli; npx skills add pascalorg/editor --skill pascal-3d --skill furniture-fit.

pascalorg /

editor

Local-first 3D building editor you can use in a browser, a CLI, or through AI agents

25.1k stars
542 stars since 3 Oct
TypeScript
Updated today
Some experience
Design & UI

Open on GitHub (opens in a new tab) Website (opens in a new tab)

Pascal lets you model buildings in 3D: sites, buildings, levels, walls, slabs, ceilings, roofs, zones, and items such as doors, windows and lights. You can use it in the browser, or install a CLI that starts the editor and keeps projects in a local database on your computer.

Under the hood, the scene is a flat dictionary of typed nodes held in a Zustand store, with undo and redo. React components render each node type as Three.js objects, and separate systems update the geometry. The code is split into npm packages (core, viewer, editor, nodes, cli, mcp), so you can embed the viewer in your own React app.

The agent side is what sets it apart. A local MCP server exposes scene tools, resources and prompts to MCP-compatible AI hosts, and the repo ships agent skills and plugins for Claude Code and Codex. The README also shows a hosted 'Pascal Next' demo, but says that experience is not yet part of the open-source release.

What it does

Local CLI with persistent projects
One npx command starts the editor and an authenticated MCP service on loopback ports. Projects are stored in ~/.pascal/data/pascal.db.
MCP server for AI agents
The mcp package exposes scene tools, resources, prompts and local storage to MCP hosts. Running pascal mcp connect links an agent without needing the editor process.
Agent skills and plugins
Two skills, pascal-3d (MCP setup and verified scene work) and furniture-fit (a footprint assessment), install through skills.sh, a Claude Code plugin, or a Codex plugin.
Reusable viewer and editor packages
@pascal-app/core, viewer, editor and nodes are published on npm. You can load the built-in plugin and mount a Viewer in a React app.
Structured scene model
Nodes form a hierarchy from Site down to Level, Wall, Slab, Roof, Zone and Items, with 2D guides and 3D scans as references. Undo history keeps 50 steps.

What you could build

Sketch a floor plan in 3D and keep the project saved locally on your machine.
Connect Claude Code or Codex to Pascal and ask the agent to build or modify a building scene through MCP.
Check whether a piece of furniture fits a room's footprint using the furniture-fit skill.
Embed the 3D viewer in your own React or Next.js app to display building models.
Study a working example of a React Three Fiber editor built on Zustand and an MCP server.

Get started

With Node.js 22.13 or newer, run the npx command to start the editor locally. To connect an agent, install the CLI globally and point your agent at pascal mcp connect, or install the Claude Code plugin.

npx @pascal-app/cli editor

npm install --global @pascal-app/cli

npx skills add pascalorg/editor --skill pascal-3d --skill furniture-fit

Commands are copied from the project's README. Read it on GitHub for the full setup.