Marek Minor, founder of Amsterdam studio Minor Adventures, spent about a year drawing Cursor’s 600+ icons in two sizes and two styles, every exploration by hand. The rule I keep thinking about: ​“Cursor’s pointer runs from bottom left to top right, and so does every icon that could go either way.” Slashes run the opposite way, because a slash cancels a direction and should cut against it.

“Gotta be honest, the hardest part of this whole project was the symbol. (Let’s be real, when is it not?)” Mike Smith on why Calendly’s new mark ended up back at ​“a cool C”. The company doubled down on its name equity, and anything figurative sitting next to the word Calendly just makes you ask why it isn’t a calendar. Dozens of explorations to arrive at the clear answer, then one asymmetrical ripple to make it theirs.

How Grok Bot was designed for agents that persist beyond a single session — from a chat history to a Bot roster, presence, a computer of the Bot’s own, and work that starts without a prompt.

“The more prominent we made the computer, the more the product encouraged users to supervise it.” The team tried four ways to show bot’s computer: floating window, side by side, modal, and full screen. Each one pulled the user into operating the machine. The shipped design has three levels instead – a purple title-bar icon while the computer is active, a pinned preview panel, and a full-screen takeover only when the Bot asks for help.

Matt Perry, who built Framer Motion and now Motion, opens The Book of Motion with perception rather than code, on the argument that ​“motion is deeply human” and hardwired into how we register that something has changed. The book will have four parts: perception, mechanics, engineering, and practice. The intro chapter is free, the rest is paid early access, published a chapter at a time. And despite the domain, it’s deliberately not a book about the Motion library.

“I worry a little bit about function following form.” That’s Katarina Batina, VP of Design at Shopify, in the middle of a piece about how completely her team reorganized around AI tooling. Roughly 60% of Shopify’s designers now push code directly to production, up 38% since last September, with over 5,500 live prototype sites built on Quick, the internal hosting platform. Her caveat is the part worth keeping: ​“There’s so much discourse around what tools are in fashion this week versus next week. The problem is we risk losing connection with the questions we’re trying to answer and the people we are building software for.”

The skills demo is the one to watch. Rodrigo calls a slash command for a skill his teammate Tammy wrote in plain English, carrying her own design-review conventions, and the agent returns the critique as annotations right on the canvas. Team practice becomes a callable command.

Miggi rounds up ten community skills for the Figma agent, though the real news is in the first line: you can now author skills inside Figma. Ask the agent to draft one using a design frame as reference, preview it, then keep iterating in chat or open the markdown editor for manual changes. Emil Kowalski’s /find-animation-opportunities, which scans a file for places that don’t animate but should, and Jakub Krehel’s /better-interface which catches the details your interface is missing, are the ones I’d install first. 

The post above shows the skills, this one shows the plumbing. Miggi builds 3 skills: a type scale generator that also writes the variables, an image audit that drops a DPI card beside every image, and Shader Starter, which pairs neatly with the shaders update below.

Everything above is agents arriving in Figma, but this one runs the other way. Nate Parrott, the Anthropic designer behind Claude Design, put its artboard canvas inside Claude Code, so options get drawn where the code already lives. One brief, several editable artboards, and then Claude implements the one you pick. /design-sync can load your design system from a repo or your local codebase first. 

Jake Albaugh, Developer Advocate at Figma, points a fleet of agents at an entire component library, then reuses the same agents as validators: each file re-fetches from Figma and surfaces what drifted after a property rename. His real argument for Code Connect is context economy – without it the MCP server hands the agent every style value it might need, and compaction is lossy.

Admins can now mark skills, templates, libraries, and Make kits as recommended within the organization, and they show up with a checkmark and a tooltip. Skills and templates can be scoped to the whole organization or to specific workspaces.

Detach the agent chat into its own window on the desktop app, move it anywhere, keep it in view while you design. Figma expects you to be watching an agent work while you do something else, which turns the chat from a sidebar into a persistent companion window.

“MCP gives your AI reach. Skills give it a recipe.” Murphy Trueman compresses two acronyms into one sentence you can actually hold onto. MCP is access: Claude reading your real Figma file, walking every fill on a button set and finding the fourteen layers still on raw hex, all of them on hover and disabled states. A skill is the playbook it follows once it’s in there. Their rule for writing one is the third time you catch yourself explaining the same preference, write it down.

The day has come! Set opacity on a color variable using a number variable, and the color stays linked to your library. No more detaching a library color just to dim a disabled state. The number variables can be scoped for color use only, so the picker stays clean.

“When we say wrap it up, we mean Pretty and Balance.” Both are the CSS ​‘text-wrap‘ values, set on a layer, a paragraph, or a text style. ​“Balance” distributes text evenly across lines, ​“Pretty” avoids the single word stranded alone on the last line.

Auto Layout’s auto spacing gets Around and Evenly, and the old default is renamed Between. Those are ​‘space-between‘, ​‘space-around‘, and ​‘space-evenly‘ from ​‘justify-content‘. Figma renaming its own concept to match CSS is the real news here, and the text wrap entry below does the same thing.

The Config announcements filled out. Shaders can now be animated and reactive to mouse movement, both shaders and generative plugins can be published to Community, a code viewer lets you read and download the code behind any shader, and MCP read/​write access.

FigJam has had auto-routing connectors for years. Figma Design still doesn’t. Sergey Tomashevsky’s Flow Pro fills that gap – wires re-route live as you drag frames, and style presets carry across files. Free for 30 connectors, then a one-time payment.

Vector editing in Figma now works by hand rather than by node. Erase a stray mark directly instead of hunting down the points that made it, then drag a color or gradient across several closed regions in one pass.

Weave is now callable from ChatGPT, Claude, or Cursor. Ask the agent to run the brand mockup tool you built in Weave, and it finds it, runs it, and returns the result in chat.

Six small curve editor fixes shipped since Config by Jakub Świadek, designer on Figma Motion: auto-scaling graph bounds, shift-snapping handles, arrow-key nudging at 0.01, and a scrub line showing exactly where on the curve the playhead sits.

The first of four new short official tutorials on Figma Motion explains easing. The detail worth taking away is that custom easing curves can be saved as variables, reusable across your entire design system.

Two tiers of preset – basic animations move a single property, composite ones bundle several plus custom easing out of the box. Once applied, an animation shows up as a blue track on the timeline, where you drag the track itself to set delay and drag its handles to set duration.

“Path trim lets you create animated lines by controlling how much of a path is visible at specific points in time. You can use path trim to create draw-on or erase effects, handwriting animations, and looping spinners.”

Comments attach to a moment on the timeline, and a single comment can carry several timestamps, so one note can flag three separate problems in one pass.

Jamie Haberman, Education Designer at Figma, builds a kinetic type plugin with the Figma agent.

“So a good rule of thumb is that semantic tokens do not include a scale.” Donnie D’Amato gives three tests for whether a token is truly semantic: it doesn’t suggest its value, it describes a generic family of components rather than a single one, and it doesn’t depend on its neighbors. That last test is the ​“smedium” problem, and it quietly disqualifies ​‘space-sm‘ and ​‘space-md‘, which is to say most of the space scales shipping today.

His recommended token construction: ‘[purpose]-[priority]-[property]‘. Priority is optional and capped at three levels, because ​“having more than 3 levels is useless for user interpretation within the experience.” The test underneath is the useful part. If you can’t point at the element in a low-fidelity wireframe, it isn’t semantic. That’s why ​‘surface-primary-shadow‘ earns its name and ​‘surface-highest-shadow‘ doesn’t.

“Names that hold up describe the decision a designer is making about the interface. Elements get renamed, colors get rebranded. The decision persists.” Murphy Trueman, Director of Design Systems at Baseline Design, calls Donnie’s piece above one of their favorite things written on the middle tier, then turns it into a working method. Walk a wireframe and ask what a designer is deciding at each element. A ​“card” is really ​“this is an elevated surface for grouped content”, so the token becomes ​‘Surface/​Elevated‘ and it survives the rename to ​“panel” next quarter.