A Product Surface for Design Agents
The useful agent interface is not chat. It is a compact set of tools that expose document state, controlled edits and visual verification.
read → act → see
Chat is a good place to express intent. It is a poor substitute for a product API.
A design agent needs to know what is selected, how nodes relate, where the viewport is and what the rendered result looks like. It also needs a constrained way to change content without corrupting the document.
Four tools cover the core loop
Fanta’s native bridge is deliberately small. One tool reads design state. One applies typed edits. One captures a screenshot. One places a generated image with its provenance.
A compact interface is easier to reason about than hundreds of UI-level actions. It gives the agent the semantics of the product without asking it to imitate every mouse movement.
- design_state: read tree, selection and properties.
- design_edit: validate and commit an atomic batch.
- design_screenshot: inspect the rendered outcome.
- place_generation: add an image bitmap with provenance.
Operations are better than clicks
A click is ambiguous without the current layout and focus state. A typed operation can express exactly which node and property should change.
The human still uses direct manipulation because it is fast and expressive. The agent gets typed, reviewable content operations. Both interfaces meet in the same document history.
An opt-in MCP surface serves external agents
The in-product agent uses four native tools directly. A separate opt-in local MCP server exposes five scoped tools to compatible coding agents over a private Unix socket.
Both surfaces target the same document, but their contracts are intentionally distinct: the built-in agent gets product-native actions, while external agents get scoped inspection and batch design tools.
Agent-ready is an architectural property
Adding chat after the editor is built can produce useful commands, but deep agency depends on the underlying model. Stable identity, transactions, screenshots and undo have to be designed together.
The interface is only the visible tip. The real work is making every action safe enough that a human can delegate without surrendering control.
The best design-agent API is small, semantic and built on the exact document the human is editing.
The takeaway