tldraw launches an offline desktop version: infinite canvas and AI agents move to the local workstation

What happened

tldraw has released the offline desktop app tldraw offline, bringing its infinite canvas from the browser to macOS, Windows and Linux. The app is built with a local-first approach: users do not need to sign up for an account, and they do not have to rely on remote servers, while still being able to draw, sketch diagrams, and edit whiteboard documents even without an Internet connection.

The app is built on top of the tldraw SDK. According to the official introduction, the same SDK is also used by tldraw.com and hundreds of other companies that are building canvas-based products.

Why this matters

Documents are stored as portable local files

tldraw offline uses standalone .tldraw files to store documents. These files can include canvas content, pages, media assets such as images and videos, and any embedded document scripts.

The official documentation defines .tldraw as the app’s native and preferred file format. Files can be shared across macOS, Windows and Linux; users can also back them up themselves, store them privately, or share them directly with others.

That makes tldraw offline more than just an offline whiteboard app; it also provides a way to manage documents without depending on accounts or cloud servers.

AI agents can participate directly in canvas creation

Another important capability of tldraw offline is that AI coding agents can view and edit the canvas. The agents named in the official documentation include Claude Code, Codex, Pi and OpenCode.

These agents can read the current document, modify content in the editor, and write reusable scripts for the canvas. The project description on GitHub also says the desktop app runs a local HTTP service and provides a Canvas API for programmatic access to tldraw documents, supporting AI coding assistants and other tool integrations.

This means AI is no longer limited to generating text or code in a chat window; it can also step into a specific visual workspace. Users can first sketch charts, process diagrams, or interface structures on the canvas, then let the agent read the canvas state and continue editing.

Evidence and confirmed information

• tldraw offline supports macOS, Windows and Linux.
• The app can be used offline without an account or remote server.
• Documents are stored as standalone .tldraw files and can be shared across the platforms above.
• Agents such as Claude Code, Codex, Pi and OpenCode can read and edit the canvas, as well as write document scripts.
• The GitHub project description mentions a local HTTP service and a Canvas API for programmatic access to tldraw documents.
• The official announcement notes that agents granted app access can read and edit user documents; scripts inside .tldraw files can also run when the file is opened. Therefore, users should only grant access to trusted agents and should only open files from trusted sources.

What to watch

First is the boundary of AI agent permissions. Because agents can read and edit local documents directly, and embedded scripts can also run when files are opened, users need to manage agent permissions, file provenance, and local backups carefully.

Second is the difference between the desktop version and tldraw.com in terms of features, collaboration, and document compatibility. The current documentation confirms offline use, local files, and AI agent capabilities, but it does not yet provide full performance benchmarks or an official release schedule.

In addition, the tldraw community has already seen adjacent practices around AI and local canvas workflows. For example, Agent Draw lets users select an area on the canvas and describe a request by voice, after which an AI agent draws a chart or illustration; other projects experiment with controlling the canvas through the command line or JSON-RPC. These projects are not official features of tldraw offline, but they show that tldraw’s canvas model is being used to build local AI workflows.

Overall, the core change in tldraw offline is not just that it can be used without the Internet, but that it combines canvas, portable files, and AI agents into a local work environment. Whether it becomes a mainstream tool for workflow design, product prototyping, architecture diagrams, and visual brainstorming will depend on agent integration, cross-platform consistency, and the real-world safety mechanisms for scripts.

Source

Official tldraw offline website