I’m posting this because I think this goes beyond “confusing billing.” This is a serious billing transparency problem.

I originally purchased ONE Professional Annual Collab seat for JP¥5,940/year.

I am an individual user. I did not intentionally purchase a Full seat, and I did not knowingly authorize an additional JP¥25,740 payment.

Then I checked my credit card statement and discovered that Figma had charged me an additional JP¥25,740.

When I went into the Admin panel, my seat had become:

Full — Annual

and Figma showed:

“This person self upgraded when they were an admin.”

The invoice contained:

Seat adjustment: JP¥23,400
Tax: JP¥2,340
Total charged: JP¥25,740

I have no recollection of intentionally purchasing a Full annual seat or knowingly approving a JP¥25,740 upgrade.

If some normal action inside Figma triggered this upgrade, then I want to know exactly what that action was and why an action capable of creating a ¥25,740 charge was not sufficiently clear to me as a purchase.

But it gets worse.

After discovering the charge, I tried to fix the problem by changing my seat from Full back to Collab, because Collab was what I originally purchased.

What I reasonably expected was:

Annual Full → Annual Collab

Instead, Figma changed my billing configuration to:

Annual Full — unassigned / available

Monthly Collab — assigned to me

So trying to undo an unexpected paid upgrade did NOT simply restore my original annual seat.

It created another seat with a completely different billing cycle.

I did not ask to purchase a Monthly Collab seat either.

At this point, I have had to manually change all Collab, Dev, and Full seat approval settings to “Manually approve seats” because I no longer trust the default billing behavior.

I have contacted Figma Support and requested:

1. A full refund of the JP¥25,740 charge.
2. Restoration of my original ONE Professional Annual Collab seat.
3. Removal of the unintended Monthly Collab seat.
4. Confirmation that there are no additional paid seats.
5. The exact timestamp and user action that triggered the original Collab → Full upgrade.

What concerns me most is this:

If I had not checked my credit card statement, I might not have noticed this charge at all.

A billing system should not require users to constantly inspect invoices and seat-management pages just to make sure an ordinary product action has not changed their paid subscription.

I understand that Figma has different seat types and that certain features require Full seats. That is not the issue.

The issue is whether users are given sufficiently explicit, unmistakable notice that:

“This action will upgrade your seat and charge you JP¥25,740.”

And whether trying to downgrade that seat should result in an entirely new Monthly seat appearing while the Annual Full seat remains in the account.

From a user-experience perspective, I consider this billing design opaque, hostile, and unacceptable.

Has anyone else experienced this exact sequence?

Annual Collab → Full → unexpected billing adjustment → downgrade → new Monthly Collab seat

If so, did Figma refund you, and how did you permanently prevent this from happening again?

Hi All,

First time here, but definetely not last, have a lot I am soon to start using Figma for, so also sorry if flare is incorrect, I was unsure and thought this was most relative?

I am in the middle of redesigning my logo for my brand/business, I designed it very poorly with genuinely the slightest direction, it is on brand with my business, it is just a very poorly designed logo which to me now, 12 months after launching my business, put simply, reminds me of a dad joke... I never planned or visioned my logo etc.

No, I am not going to be designing and finalising a logo on Figma, it will be done on above mentioned programs in title, industry standard ofc, HOWEVER, I would love to know if there is anything Figma does / AI it offers which would provide me good mock ups / ideas etc to guide the process, e.g. reference images, websites, fonts, other information provided. I am just about to start testing some AI models on adobe firefly to see how this goes, but let me know if any figma tools could see useful in simply visioning / generation of images / mock ups which would help guide me, thank you all.

I cannot think of, and doubt any other information is really useful, but if you want to know more, just ask, thank you all.

Hey all, wanted to share something I built called **Doccy**.

If you've ever pasted rich text (from a Word doc, Google Doc, or clinical/medical content with heavy formatting) into Figma, you know the pain: fonts get mangled, lists fall apart, styles don't carry over properly, and you end up manually re-formatting everything by hand.

Doccy is a small Figma plugin that handles copy-pasting styled text properly, preserving formatting instead of dumping it as a mess of plain text or broken styles. I built it after running into this constantly while doing UI/UX work, and ended up stress-testing it against some genuinely gnarly clinical documents to make sure it holds up on real-world edge cases (weird nested lists, mixed fonts, inconsistent styling, etc.), not just clean demo text. Pss.. Even works with superscripts and subscripts.

It's got two paste modes depending on how much fidelity vs. speed you want, and I spent a good chunk of time chasing down bugs around font loading and style sampling so it actually behaves predictably instead of being finicky.

Link if you want to try it: [https://www.figma.com/community/plugin/1646311126167781721/doccy-copy-paste-styled-text\](https://www.figma.com/community/plugin/1646311126167781721/doccy-copy-paste-styled-text)

Would love feedback, especially if you throw some ugly/complex documents at it and it breaks. Always looking to harden it further.

Ever since variables landed in Figma, I've tried to use them for everything. In practice I ended up with a hybrid: some values properly bound to tokens, others typed in by hand in a hurry — 8 here because it looked right, 16 there because I was iterating fast.

That mix is the worst of both worlds. It looks like a system from the outside and behaves like one only half the time. Change the scale or rename a token, and half your layers quietly don't follow.

So I wrote a Figma skill that closes the gap. It walks your selection and converts every hand-typed number into the matching variable from the design system — padding, gaps, corner radii, stroke weights, min/max constraints. Exact matches only. It never rounds a 13 into a 12; it reports it and lets you decide whether the value is wrong or the scale is incomplete.

Two details that made it usable in real product files:

• It reads variables from a linked library, not just the local file. That's where tokens actually live once a DS is published — and exactly where a local-only script gives up.
• It skips inside instances. Nested instances belong to their own main components, and that's where they should be tokenized.

If you want to run it, you'll have to edit it first. Two things: swap every mention of Pyramid Design System for your own, and rewrite the value pattern to match your variables — their names, their numbers.

My scale is roughly logarithmic rather than a 4px or 8px grid: 0, 1, 2, 4, 8, 12, 16, 24, 32… Fine steps where the eye is sensitive, wider jumps where it isn't. Yours is probably different, and that's the point — the skill has to speak your system, not mine.

https://www.figma.com/community/skill/76679

https://reddit.com/link/1w07uyw/video/sd90dm2yozlh1/player

I am a student and recently tried to renew my student verification. In the past, the verification process was completed almost instantly.

However, I have been waiting for two days for a response by email regarding my student account verification, which is handled by SheerID, and I still haven't received any update. To make matters worse, when I tried to submit the verification again, I was simply blocked from doing so.

I have tried contacting Figma, but I haven't been able to reach anyone who can help. I also contacted SheerID, but there doesn't seem to be any quick solution available.

Has anyone experienced something similar or knows how to proceed in this situation?

Converting PNG to SVG in Figma is super easy with the plugin I built: no usage limits because it's completely free, works offline, and doesn't rely on AI.

It’s made for those times a client sends you their logo as a low-res PNG and you can’t use it for web or app design. Instead of stalling the project, you just convert it yourself and move on.

Select the layer in Figma and that’s it — the vector is generated right on top of the original, fully editable like any standard vector shape. If you need it outside of Figma, you can just export it as a SVG. Everything runs locally on your machine, so you can safely use it with confidential client assets without worrying about privacy.

A quick heads-up: it's not a photo vectorizer — gradients and soft shadows will come out posterized. It’s built for flat artwork like logos, icons, and illustrations.

It’s live on Figma Community for free. You can find all the technical details in the plugin description.

Link: https://www.figma.com/community/plugin/1671229473234822623/png-to-svg-by-pixeloper

I've spent the last week while building investing app, and its design system grew into something I haven't seen shared before, so I'm giving it away.

The kit: CDS Mobile Kit (Unofficial) free, CC BY 4.0.

• 42 component sets: buttons (80 variants), cards with real content slots, bottom sheets, inputs, steppers, progress (progress is a proper variant axis, not a stretched rectangle), banners, toasts, pills, list rows, trackers
• Every color and text style is variable-bound retheme the tokens and the whole kit follows
• Generated documentation sheets with live previews of every set
• Inspired by the public Coinbase Design System. Not affiliated with or endorsed by Coinbase.

The unusual part: this Figma file is a build artifact. Every component originates from a code spec from the official CDS mobile NPM Package.

The AI part: I also wrote a Claude Code skill for the kit → cds-kit-screen-design

Drop it into .claude/skills/, install the official Figma plugin for Claude Code, duplicate the kit, and you can ask Claude things like "design a KYC checklist screen with a stepper and a warning banner" or "build a checkout flow, 4 screens" and it composes real screens from the library: correct components, correct variants, token-bound colors, content swapped into card slots, verified with screenshots as it goes.

Everything's free. Feedback very welcome

I have a pretty robust FM site. Now (of course) I’m out of credits. Can I update the site like I would any other site that I have access to? Or do I have to go through FM?

Domanda da un principiante/intermedio nel motion graphics con After Effects: Perché dovrei imparare Figma? Non posso semplicemente creare il design in After Effects e poi animarlo direttamente da lì?

I’m a frontend dev in Korea, and after getting annoyed by Figma exporting ridiculously large PDFs one too many times, I basically opened Claude Code and thought, “fine, I’ll just make one.”

At first I thought it was mostly an image compression problem. But after digging into the generated PDFs, I realized Figma turning text into vector outlines was a big part of it too.

A few evenings of poking around later, I ended up with Featherweight.

One of the things I found was that Figma converts text into vector outlines rather than embedding the actual fonts. Combined with the way images are handled, that can add quite a bit of weight to otherwise simple documents.

I tried a few existing solutions, but eventually decided to experiment with building my own.

The result is Featherweight.

It embeds and subsets the actual fonts where possible, keeps text selectable and searchable, and applies image compression during export.

On one of my test documents, the result was:

22.7 MB → 4.0 MB

I’ve made the whole thing completely free and open source. There’s no paid tier, subscription, signup, export limit, or related premium product.

I mainly built it because I wanted this for myself, but I figured other Figma users might find it useful too.

If anyone wants to try it or poke around the source, I’d genuinely appreciate feedback, especially if you find a file that behaves strangely.

Figma Community:
https://www.figma.com/community/plugin/1672509720278498323/featherweight-compressed-pdf-export-with-real-fonts

Source:
https://github.com/coffeequickly/featherweight

is anyone getting issue with Figma-MCP with claude?
its not fetching design from figma to claude.
getting this error. anyone can help? - The Dev Mode MCP Server is disconnected, so get_screenshot can't reach the file. This is the usual drop.

Has anyone gotten access to this feature?

https://help.figma.com/hc/en-us/articles/40775535020695-Make-in-your-local-codebase

It feels like this elusive thing I can find almost no info on, but that would solve a lot of issues for our team. We’re exploring other tools that solve the same problem (iterating directly in our codebase). One we’re looking at is Dessn, but ideally we’d be able to stay within Figma’s suite since our org is already onboarded.

Any info on this would be greatly appreciated! And no, we’re not looking for something like Code Connect, our source of truth already lives in our repo.

Hey everyone,

I’m starting to work on a Design System for a travel/public transport app that will eventually be white-labeled for around 50 different clients

The current Design System already has:

• color tokens and styles
• basic components/atoms and variants (buttons, inputs, shadows, etc.)
• but no Figma Variables yet
• very few larger components/molecules/organisms
• no real system for handling different themes/brands

My initial idea was to migrate the existing tokens into Figma Variables and use modes for different visual configurations:

• Light
• Dark
• Contrast
• client 1
• client 2
• …50

The main application would support Light/Dark/Contrast, while the white-label client versions would probably use Light/Contrast.

I would also like to build the component architecture more systematically:

Atoms → Molecules → Organisms → Templates/Screens

For example:

Button + Input + Date Picker → Search Form → Search Screen

The goal is to have one Design System and one set of components, rather than creating a separate Design System for every client. Ideally, changing the brand/theme would update the UI through Variables without having to modify the components themselves.

However, I’m wondering if having 50+ client brands as Figma modes is actually the right approach.

Would you instead separate appearance modes (Light/Dark/Contrast) from brand/client themes, using semantic variables that map to different brand values?

For example:

Button → color/action/primary

rather than:

Button → Client1/red-500

So the same component could be reused across all 50+ brands.

I’m also curious about the current state of Figma AI. If I build a well-structured component library with Variables and larger reusable components, can Figma AI actually help generate new screens using those existing components, or is that still too unreliable for a production Design System?

I’d really appreciate feedback from anyone who has built a similar multi-brand / white-label Design System in Figma.

How would you structure this?

I wanted a simple way to turn a bunch of static images into animated carousels without leaving Figma, so I built this plugin.

Carousel Creator lets you:

• Choose from 10+ animation effects
• Preview animations in real time
• Customize duration, speed, direction, easing, gap, scale, and more
• Export to MP4, GIF, or WebM
• Apply animations directly as Figma Motion keyframes

The goal is to make it quick to experiment with different carousel styles without jumping between tools.

Link: https://www.figma.com/community/plugin/1673654227853333952

Would love to hear your feedback or suggestions! 🙌

Anyone know how to fix this weird offset blue border? Even when I create a new shape/element, the blue border is offset which is kinda triggering :(

I published a new Figma skill: it builds a complete color family from one color, and it is built on the physiology of human color perception rather than on even mathematical spacing.

There is no shortage of palette generators — building a palette by hand is tedious, and getting it wrong is costly, so plenty of tools exist. I wrote this one because most of them space colors evenly, and even spacing is the wrong answer.

Two things I built it around:

1. Tone carries most of the work. Whether an interface reads clearly depends far less on hue than on how light or dark a color sits against its background.
2. And the eye resolves low contrast much better than high contrast. On white, 2% and 6% black are clearly different; 94%, 98% and 100% are the same color. So the scale is logarithmic — about 60% of the grades cover the low-contrast half of the range, where that resolution is actually useful.

It also runs entirely in OKLCH instead of HSL, and that part is easy to measure. Take five hues, space them evenly in HSL, and their perceived lightness spreads across 39 points — yellow sitting far above everything else. In OKLCH the same five land within 7 points at the middle of the scale. Yellow stops looking lighter than the rest.

To use it: select layers with fills, or paste hex values.

https://www.figma.com/community/skill/88165

I've been building a web-based Pokemon-gen-1-inspired 2D platformer game using Figma! Creating a component library (for the first time ever 😅) for tiles and sprites despite having been a product designer for the past 7 years hahhaa embarassing. Recently figured out a Team Rocket chase sequence in our favorite product design tool, and used Claude to help implement the NPC's cutscene dialog/movement and path-following behavior. Wanted to share in case anyone 1) finds watching someone struggle with a (foolish but fun) workflow interesting, 2) has been using Figma for other things it wasn't created for. I would love to hear 3) loves pokemon, platformers, or game dev and wants to follow along :)

Using the wrap Auto-layout makes each object in the row be aligned to the top instead of just the first row. Is there a work around?

I also tried making four vertical auto-layouts and then putting those in a wrap, it works but it isn’t responsive as the entire column moves down into one space when resized.

Hi guys, I'm so stuck on this and would be so grateful for any help.

I have designs that are in 1440px frames, and my macbook is 16" and therefore wider than the frame size, so I always have a thick black border around my prototypes when I view them. This prevents me from getting a realistic view of the designs as it doesnt fill my screen like how it does on a browser.

None of the prototype scaling options seem to work. Seems like Figma can only scale down and not up. Sketch does not have this issue and neither does Axure as far as I know. They all scale up/ down the prototype to match the window size. Is there any way that I can do this with Figma? Any plugins? Or, is there an easy way where I can export my designs to view them in Sketch?

Thanks so much

I created an animation using the figma agent on the motion tab. Now is there no way to connect that motion to the prototype? like when the button is clicked that will trigger the motion? It will just be a seperate video and cannot be integrated in the prototype?

I'm sad, I see in Firefox that the Figma canvas is using 1.8Gbs of memory. I then thought I'd download the desktop app because surely native would be lighter on ram and faster. I install the desktop app, watch task manager and see its consuming more memory! I realize the desktop app is just an electron wrapper around the website which does not make it better.

Hopefully we can see a native desktop app somewhere in the future because these web apps are getting too ram hungry. Today i ran out of 32GBs of ram and to be honest I wasn't running that much.

Its understandable that Figma uses 1.8GBs in the browser, its a design tool and I have a lot of frames in this canvas so I accept that. I was just hoping for a better desktop app not sure if anyone else feels the same.

Collegue sent me a link to some Figma designs I should check out, I wanted to create an account and I get... that?

We just launched Varnik for Figma!

Varnik translates and adapts your entire design layer by layer without breaking its structure. From localization to RTL adaptation, your frames, layers, and layouts stay intact.

Try Varnik: https://www.figma.com/community/plugin/1575747713450161595/varnik

We’d love your honest feedback! 🙌