Atriana

AI-native architectural design

Atriana replaces traditional CAD workflows with an AI-first approach. Instead of drawing lines and shapes, you describe what you want to an agent, which builds and modifies the design through Atriana's MCP server. The browser UI is a live view onto canonical state held by a local Node server, kept in sync over WebSocket.

Designs are represented as semantic objects (walls, shelves, panels, flue pipes, granite supports) rather than raw geometric primitives. The agent reasons about what it's building, not just where lines go, and the app feeds screenshots back so the agent can see its own work and iterate.

You drive Atriana from any MCP-capable agent: Claude Code, Cursor, Claude Desktop, or anything else that speaks MCP. Point it at the local endpoint and start prompting.

Example: modeling a wood-burning stove

I gave Claude a photo of an existing wood stove and asked it to recreate the design through the MCP. A few seconds later the model was sitting in Atriana, ready to inspect, edit, and export.

A real wood-burning stove with wood storage at the base, used as the reference for the design — The reference photo passed to the agent.

Claude Code prompt: 'can you use the mcp to generate this fireplace in a new doc' with the reference image attached — The full prompt. No CAD vocabulary, no coordinates, just the photo and an instruction.

Atriana app showing the wood-burning stove design in 3D rendered view, with the MCP log on the left and a components panel on the right — The 3D result. The MCP log on the left captures every design call the agent made; the components panel on the right lists the semantic objects (flue pipe, granite base, stove door glass, wood panel back wall, and so on).

Atriana app in 2D drafting mode showing top, front, and right projections of the same wood-burning stove design — Same canonical state in 2D drafting mode. Top, front, and right projections are generated automatically and can be exported as PDF, SVG, or DXF.