Figma Release Notes

NEW RELEASE

FIGMA MAKE

Skills are markdown files that outline the conventions and workflows you use repeatedly, so you get prototypes that match your standards with fewer prompts. Import existing skills or create one in Make, then call it with a slash command in any prompt.

Use a skill to bring in context you want reusable across all your Make files, like /insert-sample-data for dropping in company-approved test data. Or use one to run a repeated workflow the same way every time, like /build-from-prd paired with a Notion or Confluence connector to turn any PRD into a prototype that meets your standards.

Today, each person creates and manages their own skills. You'll soon be able to publish and share skills across your team and organization.

Learn more about custom skills in Make.

NEW RELEASE

UPDATE

DESKTOP APP

FIGMA

Move between files in the Figma desktop app without the usual interruptions that slow you down. Links open where you already are, recent work is a search away, and what's next loads before you get there.

• Search open and recently closed tabs by name from the tab overflow menu
• Open Figma links directly in the desktop app, skipping the browser routing step (macOS)
• Files and prototypes preload in the background so they're ready when you arrive

NEW RELEASE

UPDATE

FIGMA MAKE

DESIGN

PROTOTYPING

Make introduces new features for more control and context as you build:

• Voice-to-text - Dictate prompts directly in Make chat with cleaned text ready for review before submitting. Learn more about voice-to-text.
• Question cards - Choose between structured options in chat before Make moves forward, each with a short description of the tradeoff. Learn how to use question cards.
• Version history - Track changes across every version of your build and revert to any prior state instantly. Learn how to use version history.
• Clear context - Reset your session in one click from Make settings, clearing accumulated context without losing your file or your work. Learn how to clear context.
• New connector - Bring in additional context from Google Drive, Microsoft Office, Zoom, and 9,000+ other apps with our new Zapier connector.

UPDATE

FIGMA
FIGJAM
FIGMA SLIDES
FIGMA BUZZ
FIGMA DRAW
DESIGN

It’s now easier to add references when using Make Image and Edit Image in Figma. Previously reference images could only be uploaded, but now you can:

• Click the new "Add reference" button on almost any node on the canvas (images, vectors, frames, components)
• Copy/paste into the prompt box (from the canvas or outside Figma)
• Drag and drop files into the prompt box

Available in all Figma products where you can make or edit images with AI: Figma Design, Draw, Buzz, Slides, and FigJam.

NEW RELEASE

UPDATE

FIGMA

FIGMA DRAW

Draw now surfaces the tools designers reach for most without leaving the mode they're already working in. New expressive controls expand what's possible natively, alongside the core features that were always there but harder to find.

Features that were always there are now easier to find and use.

• Apply auto layout directly in Draw, without switching to Design mode
• See component, instance, and text layer types labeled inline in the layers panel
• Use the new dedicated text on a path tool to add text to an existing path, or drag on an empty canvas to create text on a circle
• Right-click to separate text and vector into independent layers

New brush and texture controls bring more precision and creative range to Draw.

• Cmd-click or ctrl-click any stroke to sample its color, weight, and brush or dynamic style
• Set a gradient color and blend mode from the Draw toolbelt before starting a stroke
• Control X and Y noise and texture independently to create directional effects

Learn how to type text on a path in the help center.

Agents are changing your code faster than your team can follow. Now you can close that gap with new MCP skills, architecture layouts, and more in FigJam.

As a backend engineer, I usually live in the infrastructure layer—building platforms, designing architecture, pitching a different technical approach. I've never touched the C++ editor or made a single edit to Figma Design. But during a recent MCP team hackathon, I found myself doing something I'd never done before: advocating for another team's product, FigJam. Specifically, making it the place where engineers and agents think through systems together.

My team is shipping faster than ever, building features that would have taken quarters in weeks. But there's a hidden toll. Every agent-written PR we merge without deep human review adds confusion and hidden complexities to the codebase. As a visual learner, I wanted a way to zoom out and see the bigger picture of all these changes.

So during the hackathon, I built an extension of our existing generate_diagram tool to support more complex architecture and ERD layouts. From there, my team created figma-use-figjam—a new MCP skill that lets agents read and write directly to FigJam boards—and workflow skills like generate-project-plan that turn docs, codebases, and conversations into visual boards. One place for the whole team, agents included.

These tools and skills all launch today. They build on use_figma, an MCP tool we shipped last month that lets AI agents create or edit designs directly on the Figma canvas using your actual components, and create_new_file, which allows agents to generate designs inside new Figma files. Together, these capabilities let any team recreate this kind of workflow in FigJam. Here's how it works:

Step 1: Research, plan, and make it visual

First, I have my coding agent fetch the relevant context for planning the addition of a new tool—MCP server documentation, codebase structure, existing patterns. Then, the agent identifies potential solutions and researches tradeoffs. I have the agent recap what services or files are affected. Then, I map subtasks to stacked PRs and design a testing strategy.

At this point, I have solid implementation options, but they’re hidden in a wall of markdown—not the easiest thing to drop in Slack for feedback. Now I can use the new FigJam capabilities to turn that plan into an interactive eng crit in minutes:

• Architecture and ER diagrams from the generate_diagram MCP tool and skill
• Notes, code blocks, and annotations from the figma-use-figjam skill

What used to be a dense block of text is now a visual my team can actually engage with—and the cleanest architectural approach becomes obvious.

Step 2: Collaborate before committing to code

Next, I share the board with my team. Teammates jump in to leave feedback and ask questions: Will this tool support multiple file types, or just design files? Should it accept a folderId, or create all new files inside the user's Drafts folder?

We may not all be in the same conference room whiteboarding with a marker anymore, but FigJam brings all the technical context into a familiar, collaborative visual format. With these new integrations, we can brainstorm and collect feedback on agent-generated diagrams across remote teams.

Step 3: Start implementing

Once all the reviews are in, I take the plan back to my coding agent, iterate based on the comments, and get ready to implement. Previously, that meant screenshotting the diagram, recapping the comments, and manually explaining what changed. Now, I use the get_figjam tool to grab everything from the board—the diagram, the decisions—and bring that into my coding environment to start implementation. By the time I put the PR up (which links to the Figjam with all the design context, of course), it's an easy merge because the architecture has already been reviewed.

Build on this foundation

Seeing your systems mapped out has always mattered—but in this new era of development, it matters even more. And once something's in production, keeping those diagrams up to date manually is its own kind of work. I originally built this to visualize my own understanding at the speed of code. Since then, I've watched it take on a life of its own at Figma—developers, PMs, and cross-functional teams are using it for everything from onboarding docs and eng crits to tech specs and PRDs.

The tools and skills we’re shipping today are intended to be building blocks—a foundation you can use to support your own development process. Use generate-project-plan to turn a spec into a visual board you can collaborate on with your team. Build your own skills on top of figma-use-figjam and figma-generate-diagram to fit the way your team actually works.

Read our guide to the Figma MCP server to start using this tool, or explore our developer docs to see what's possible when building on top of figma-use-figjam. You can also contribute skills by submitting them to our GitHub repository.

This will be a paid API, but it’s free during the beta period as we learn how to account for agentic behavior in our paid seats. It currently works with MCP clients including Augment, Claude Code, Codex CLI, Copilot CLI, Copilot in VS Code, Cursor, Factory, Kiro, and Warp.

Caroline Okun is a software engineer at Figma working on AI integrations and developer tooling. She’s led major infrastructure efforts across search, notifications, and realtime systems, including redesigning core platforms and most recently launching the Figma MCP server.

NEW RELEASE

FIGJAM

Table formatting, diagramming, canvas navigation, and template publishing in FigJam just got a bit better with the following updates:

• Cell merging: Merge adjacent cells while preserving content from the upper-left cell.
• Text color in tables: Apply distinct text colors within table cells or shapes.
• Improved arrows: Wider routing margins, cleaner arrowheads, and clearer dashed endpoints.
• Drag-to-flip shapes: Drag any resize handle across a shape to flip it. Content stays readable.
• Recenter button: Quickly find your way back to your work on large canvases.
• Default zoom decrease: Start slightly more zoomed out for better board visibility.
• Template publishing on Professional plans: Publish up to 5 templates per team.

Learn more about using FigJam.

NEW RELEASE

FIGJAM
AI
DEVELOPMENT
DIAGRAMMING
MCP
MEETINGS & WORKSHOPS

With new MCP tool updates and FigJam skills, you can now generate architecture diagrams, ERDs, and more in FigJam directly from your coding agent — including into files you're already working in.

• Generate diagrams: Use the updated generate_diagram tool (remote server only) to create architecture diagrams and ERDs, with new connector types built for database relationships. Or paste Mermaid.js code directly onto the canvas to render it as a diagram.
• Use new skills: Read and write directly to a FigJam board with the figma-use-figjam foundational skill. Workflow skills like generate-project-plan turn docs, codebases, and conversations into visual boards.
• Get context: Use get_figjam to summarize an existing FigJam board, surface insights, or draft next steps.

Available in MCP-supported clients. Current pricing and rate limits apply.

Explore how FigJam's new MCP skills help you visualize the code your agent is writing.

Learn more about using MCP tools in our help center.

UPDATE

FIGMA

DEVELOPMENT

Plan Access Tokens (PLANTs) are now in beta — a new authentication method for server-to-server integrations and internal automation. Tokens are scoped to a plan rather than a user, so they remain active even if the token creator's account is deactivated.

Available to admins on Organization and Enterprise plans.

• Better security and governance — plan-scoped access with admin controls over token creation and usage
• Improved reliability — tokens survive personnel changes, keeping automations running
• Enhanced visibility — clear audit trails showing which tokens are accessing your data
• Higher rate limits — better suited for high-scale automation workloads

Learn more in developer docs.

NEW RELEASE

UPDATE

MOBILE APPS

FIGMA

FIGMA MAKE

PROTOTYPING

Make is now available in the Figma mobile app, bringing mobile testing and review to your device. Preview a Make on your phone, experience it with real gestures and constraints, and share for review on iOS and Android.

• Tap through flows and validate scrolling, spacing, and interactions
• Share Makes with collaborators for review and approval

Available in the latest version of the app, rolling out over the next week. Update or download the Figma mobile app to get started, or learn more about Make on the Figma mobile app.

NEW RELEASE

FIGMA

FIGJAM

FIGMA SLIDES

FIGMA BUZZ

DESIGN

OpenAI's newest image model, ChatGPT Images 2.0, is now available to use with Make Image and Edit Image in Figma Design, Draw, Slides, Buzz, and FigJam, and also in Figma Weave. ChatGPT Images 2.0 reliably generates high-quality images, and excels at producing smarter visuals like infographics or multilingual generations, executing on better editing and aesthetics, and preserving faces across generations.

Learn more in the help center.