Published 10 September 2026 · Source dated 7 May 2026
Figma's design agent moved out of chat and onto the actual canvas, working with skills, web search and MCP connections. It's in open beta and already wired into real team workflows, not just a demo.
At Config 2026, Figma showed its design agent working directly on the canvas: generating and remixing layers, running custom **skills** your team defines, searching the web for real context, and connecting to outside tools through **MCP** — reading data in and writing changes back out. Teams like Uber are already using MCP skills inside FigJam so the agent pulls real workflow context instead of guessing.
Why it matters for you: this touches the actual file — components, tokens, layers — not a mockup of one. Penpot has no equivalent yet, so if you split time between the two, the gap between them just widened.
Coach
Open a spare Figma file, turn on the agent beta, and give it one small task like tidying a component's auto-layout — watch exactly what it touches before you trust it on real work.
A September 2026 roundup shows AI agent tools narrowing in on single business tasks — procurement, invoicing, logistics — rather than acting as general chatbots. None of it is design-tool specific yet, but the same narrow-agent pattern is what design tools are chasing too.
pick one repetitive task in your workflow (renaming layers, writing alt text, chasing feedback) and see if a narrow AI agent — not a general chatbot — can take a first pass at it this week.
A new coding-agent platform lets you run dozens of AI agents in parallel from one dashboard. It's dev-focused today, but it's a live example of where design-to-dev handoff tooling could head next.
Skim how the '/workflows control room' concept works, then name one repetitive handoff step in your own process (spec writing, redlines) you'd want an agent to own first.
Figma now ships a design agent that works directly on the canvas, plus an MCP server that lets AI tools read your actual file data instead of screenshots. This is a real product shift, not just a demo.
open one messy Figma file, rename your layers and components properly, then try Figma's MCP server (or a Penpot plugin) with an AI tool to see what it actually reads.
