Figma Agent
Product Hunt Kate Ramakaieva
Figma launched Agent, a new canvas-native assistant for design work. It edits real files and can turn prompts, comments, and context into prototypes.
Based on reporting by Product Hunt, Kate Ramakaieva — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Figma has launched Agent, and it lives inside the Design canvas instead of floating off to the side like a helper widget. That matters because it works with the actual stuff teams use: real components, variables, and shared files, not flattened images.
The pitch is pretty direct. A designer can ask it for layout directions, bulk-edit across screens, turn comments into changes, and build prototypes from a prompt. Figma is framing this as a way to stay in flow from idea to review, which is exactly the kind of promise these tools keep making — and sometimes, occasionally, even deliver.
There’s also a broader workflow play here. Agent can pull context through MCP from Notion, Slack, GitHub, and Linear. It can build custom plugins and shaders, and teams can save repeatable workflows as skills that run with “/”. That turns the product from a single AI feature into something closer to a small operating system for design work.
This is Figma’s 34th launch, which tells you the company is not treating AI as a one-off bolt-on. It’s trying to make the canvas itself feel less like a static place to arrange pixels and more like a working surface where prompts, feedback, and shipping all blur together.
My take — AI-written commentary, not fact-checked reporting
The smart move here is not the prompt magic; it’s keeping the agent inside the real workflow instead of making designers copy-paste their lives into yet another side panel. That’s the difference between a tool and a toy. Figma seems to understand that the future of AI products is often just less friction, not more spectacle.
Read more about this at: Product Hunt