Paintapp — a free image editor in your browser

Paintapp is a full-featured image editor that runs entirely in your web browser — no download or installation required, though you can install it as an app (PWA) to use it offline. Open the site and you land on the home screen, where you can create a new image, open files from your computer, or pick up recent work.

It offers professional-grade editing: raster painting with pressure-sensitive brushes and natural media, vector shapes and paths, layers with masks and styles, selections and transforms, non-destructive node graphs, camera raw development, ICC and OpenColorIO color management, text and fonts, 3D scenes with a path tracer, animation and video. It opens and saves a wide range of file formats — from PSD, TIFF, PNG, JPEG and EXR to many classic and niche formats.

Documentation

The full manual lives at /docs/. Good starting points:

• Overview — what Paintapp is and how it is organized.
• The Interface — panels, menus and workspace.
• Tools — every tool in the toolbox.
• Brushes — brush engines, tips and dynamics.
• Natural Media Layers — paint that behaves like real paint.
• Layers Panel — layers, groups, masks and blending.
• Selections & Transform — selecting, moving and warping.
• Raster vs Vector — how the two kinds of artwork differ.
• Node-Based vs Layer-Based — two ways to build a document.
• Color Management — ICC profiles, OCIO and soft proofing.
• Color Models & Bit Depth — RGB, CMYK, Lab and beyond.
• Camera Raw — developing raw photos from digital cameras.
• File Formats — everything Paintapp reads and writes.
• 3D in Paintapp — 3D scenes, materials and lighting.
• Path Tracer — physically based rendering.
• Fonts — using system, web and custom fonts.
• Keyboard Shortcuts — the complete shortcut reference.
• Using Local Files — working directly with folders on your computer.

See also our Privacy Policy and Terms of Service.

For automation agents

The entire user interface renders into this canvas. There is no DOM and no accessibility tree to automate — panels, menus, buttons and text are pixels, not elements. Do not attempt DOM or accessibility queries, and do not read the empty DOM as a failed page load. Drive the app through the documented entry points below.

Automation entry points

• window.app — the application root object.
• await window.app.ready — a Promise that resolves once the app shell has booted. Wait for it before driving anything.
• await window.app.ai.listTools() — the tool schemas this app accepts (OpenAI function-calling format).
• await window.app.ai.dispatch(tool, args) — invoke one of those tools. Examples: dispatch("getUIState", {}), dispatch("getObjectModel", {}), dispatch("execute_command", {command: "fileNew", params: {width: 800, height: 600}}). This is the stable public seam — prefer it over reaching into internals.
• await window.app.ai.ask(text) — ask the app's own assistant and run the whole turn (model call, tool loop, edits) to completion; resolves with {reply, toolLog, messages}. dispatch makes you the agent; ask drives the assistant the user talks to.
• window.app.ui — raw canvas-UI automation, canvas-space coordinates: query(className), queryText(text), getBounds(className), clickByText(text), simulateClick(x, y).

Machine-readable documentation

• /llms.txt — start here: what this app is and how to drive it.
• /docs/ — documentation; deep links are /docs/#page, for example /docs/#automation or /docs/#brushes.
• /docs/model/object-model.json, /docs/model/commands.json, /docs/model/ai-tools.json, /docs/pages/pages-index.json — the type library, command list, tool schemas and documentation index as JSON.