Aetherboard · AI workflow
Your AI can see the board. You keep control of it.
Aetherboard keeps its working material in readable folders—not behind a proprietary cloud API. A compatible folder-aware agent can understand the same board you do.
Point it at the library
Choose the Aetherboard folder on your computer. The agent can inspect board structure, text and image references as ordinary files.
Ask for useful work
Have it find a phrase, organise material, add research or prepare a visual response while the board remains open.
Receive, then decide
Agent changes arrive through a validated inbox. Aetherboard checks the shape of the change before it becomes part of the board.
An honest boundary
Aetherboard is not an AI chatbot wearing a canvas.
It does not send your boards to a model by default, and no assistant is hidden inside the app. You choose the agent, grant it folder access and decide what work it should do. The whiteboard remains useful with no AI involved at all.
Readable source
Code your AI can read or edit.
Aetherboard/
└── workspaces/
├── Work/
├── Life/
├── Learning/
└── Scratch/
└── Your board/
├── board.json
├── images/
├── inbox/
└── backups/Local first
The useful part is the openness.
Agents change. Models change. Your board should not have to. Aetherboard gives the work a stable home on your own computer.
Read the tested inbox mechanism →