Help others

I use Claude on two separate Claude accounts, and both need the Figma MCP connector pointed at the same Figma account. This does not work. As soon as I authorize the Figma connector on the second Claude account, the first one is invalidated and its Figma tools stop working (“The user’s connection to this connector was invalidated”). Re-authorizing on the first account then kicks off the second. The same Figma login can be authorized in Claude Code’s own Figma plugin at the same time without conflict, so the limit seems to be one active authorization per Figma account for the Claude connector client, rather than a general one-session rule. Please allow one Figma account to hold more than one active authorization for the same MCP client, or document the limit and the intended workaround. Reproduced today on the current remote server at mcp.figma.com with two Claude accounts, one Figma account.

Description:After today's Figma update, the option to reuse recent prompts is gone from the AI image tools (Make an image / Edit image). Before the update, I could pick a previously used prompt from a list. Now I have to retype the full prompt every time I generate or edit an image.Steps to reproduce:Select an image or frame and open the AI image generation / edit image tool. Enter a prompt and generate. Open the tool again on another image.Expected: A "Recent prompts" list (or history) is available to reuse or tweak previous prompts.Actual: No prompt history. The input field is empty, and the prompt must be written from scratch.Impact: This significantly slows down iterative work. I often apply the same or slightly modified prompts to many images in a row, and retyping them each time is a major workflow regression.Request: Please restore the recent prompts history, or let me know where it was moved if it still exists.Environment: Figma desktop app, macOS  

Hello. Since the latest UI3 updates, the native AI image generation and editing tool has suffered two major usability regressions that significantly slow down my workflow:Missing Prompt History: The "Recent prompts" history (the clock icon) is completely gone from the new floating toolbar. Every time I need to reuse or tweak a complex prompt on a new image, I am forced to retype everything from scratch. Missing Quick References (+): Before, while the image generation panel was open, I could simply click a "+" icon on any image or frame on the canvas to add it as a reference automatically. That option has been removed, and adding references now takes considerably more steps.Together, these changes make iterating on AI images much slower and more frustrating than before. Please restore the recent prompts history tab and the quick canvas reference tool as soon as possible. Thank you.

Hello,I've been using Figma for a while, and I've been able to use PS5 DualSense's PS button to trigger interactions. However, I recently updated and all of a sudden, button inputs are reading the way they show up on https://hardwaretester.com/gamepad . Now I can't use the PS button (Home) as a trigger anymore. Does anyone know how to fix this? Thanks in advance! 

I design my emails in Figma and then chop them up using slices. I created a Claude skill that automatically exports Figma elements to Klaviyo, specifically these slices. Everything is working well EXCEPT - slices for some reason do not include any Figma images elements. which means no background images, no logos, product images, nothing. If i export manually it works okay.I noticed that if I export frames however, it works as intended but switching to frames is not an option for now.Figma team, please fix MCP or API so that when exporting slices via Claude MCP it includes images as well.

I have linked my prototype frames a lot of times it was for a manual pop-over in prototype but it gets disconnected again and again automatically after few days, also for some other frames as well it happens is this some bug or something else i am not able to understand I am using web version of Figma.

Your improvements to auto layout are incredible. I can’t tell you how much I love seeing the Figma team tackle these core features. There is a “final boss” of the Grid auto-layout that I’d like to suggest potentially tackling. It looks like you’re heading in this direction, so please carry on if you’ve already started :-)“repeat(auto-fit, minmax())” Responsive grid columns, equivalent to repeat(auto-fit, minmax()).Expose a minimum card/column width in the GUI, with the grid automatically reducing its column count and adding rows as the container narrows. This would make a powerful—but challenging—CSS pattern much more accessible to designers and some developers. 

I am able to create a table using the grid auto-layout option. but I’m not able to reorder the columns or rows. It would be a great feature.

Summary: After performing a Library Swap, frames with a specific variable mode ("Light") applied are rendering the incorrect mode ("Dark"), despite the UI indicating that the correct mode is selected.Steps to Reproduce:1. Open a file containing frames with variable modes applied (e.g., a "Light" mode from a specific Variable Collection).2. Use the Swap Library feature to switch the source library to a new version/library with equivalent variables and modes.3. Observe the frames that were previously set to "Light" mode.4. Attempt to toggle the mode from "Light" to "Dark" and back to "Light" via the Layer panel.Expected Behavior: Following the library swap, frames should correctly render the variables associated with the "Light" mode. If the UI says "Light," the visual output should match.Actual Behavior: The frames render as "Dark" mode visually, even though the Layer settings panel explicitly shows "Light" as the active selection. Manually toggling the modes in the settings does not fo

I built a Figma companion plugin that allows an external hardware controller to interact with the Figma canvas (create shapes, align layers, toggle visibility, export images, etc.) via localhost HTTP.It was rejected with the reason: “not able to approve apps that overlap with native capabilities in the Figma MCP server” and suggested using the use_figma tool.However, the MCP server is not a viable alternative for this use case:OAuth is restricted to approved AI clients only (Claude, Cursor, VS Code, etc.). Custom applications cannot request the mcp:connect scope. The local desktop MCP server is read-only — use_figma is not available. The REST API cannot write to the canvas.My plugin is not an AI tool — it’s a hardware device integration. The companion plugin architecture (Plugin API + localhost HTTP) is currently the only way for a non-AI external application to write to the Figma canvas.Is there an alternative pathway for hardware integration plugins? Or is there a way to request MCP

Unsure if this would be considered a bug or feedback but it’s impossible to write words including the letter “n” within the text properties of components in Figma Buzz’s design mode tab in Asset view. The moment I type “n,” it immediately moves to the next frame instead of updating the text property or allowing me to continue typing. Editing the text directly in the canvas does not cause this issue, but it is frustrating to work around and troubleshoot for non-designer collaborators who are more comfortable using the properties panel instead of directly editing the text object.

I've been experimenting with multiple video fills in a fairly longer than average Figma Motion composition (2-3min). Thanks to the workaround from another thread, I've managed to get videos playing in the final export by adding two rotation keyframes.However, video synchronization still seems to be an issue. For example, I have a 141s video that's supposed to start at 0:00 when its layer appears at 32.14s in the composition. Instead, Figma starts playing the footage roughly 32 seconds into the original video, as though it's following the overall composition timeline rather than the individual layer's timing. It also unexpectedly loops later on randomly somewhere you don’t wanted.I've tried adjusting layer duration, opacity (0% → 100% with Hold easing), and Position/Rotation keyframes without success.Of course, Motion is still in Beta, but having independent playback controls for video fills (at least a start time, offset, trimming, or looping) would be incredibly helpful for more compl

Hi guys, I subscribed to figma not because I just wanted to use Figma Make. I thought the AI worked for designs as well, just like https:sleek.design, but unfortunately for me, what is made, can’t be edited on Figma Design. How do I cancel this subscribtion. It’s been taking almost £22 from my account monthly - no matter what I do.Many thanks.Abe

I moved to Linux ages ago and if I’m honest I am nothing but dissapointed with the lack of support. Linux needs a native application for stuff like fonts and ease of use, and the design team doesn’t seem to care. It’s been proven that Linux is eating Windows’ lunch and that adaptation is needed, do they think that Linux is so open source that people will steal their app? That’s not how that works.The only reason I use other tools such as Lunacy, Inkscape and GIMP is because Figma isn’t supported, and that fact also prevents me from spending any money on a subscription from them. How the hell can we convince these people to make AND PROPERLY MAINTAIN a native Linux application. It just has to be a Flatpak on Flathub. That’s all. But no, it looks like the company worth 11.7 Billion just can’t figure out how to wrap a LITERAL WEB APP and give it desktop features as a Flatpak… disappointing and deterring. 

For a change of 10 Variables - the Agent used +1300 Credits (+Error in first run, consuming 800 Credits)!Its a total use of +2000 Credits for - 1 Prompt - that is rather a low effort/severity. Its nearly half the amount of the monthly budget of 4250 credits.Whats your experiences? 

Hi there!A few months ago, our campany merged with another one whose designers used to work with Balsamiq. They’ve just migrated to our Figma organization workspace but obviously wouldn’t want to lose their Balsamiq ressources.Does anyone have any advice or process to share to transfer Balsamiq files and librairies to Figma ?Thx a lot for your help,Jess

This is a suggestion for a “quality of life improvement” new feature. Current behaviour:The variable selection within a collection and its extended collections are not synced.When one selects a variable in an collection / extended collection, the variable selection can be different for a collection and its extended collection.Steps to recreate current behaviour: Select collection P Select variable A within that collection Select extended collection (P extended) Select another variable B within the extended collection Switch back to collection P The selected variable in collection P is still AExpected behaviour:The expected behaviour is that the selected variable should be synced between a collection and its extended collections.Typical flow of a designer is to extend a collection and then modify specific variables within the extended collection. They switch back and forth between the collections to validate the previous variable value and the modified variable value. Steps to recreate

Hi Figma team,I’d like to report an issue, or ask whether there’s a recommended solution, related to how the Figma desktop app updates itself on Windows.What happensEvery time Figma updates, a new folder named after the version is created under C:\Users\<username>\AppData\Local\Figma\ (for example app-126.8.18, app-126.9.11, and app-126.9.13, as shown in the screenshot below). The Figma.exe that actually runs appears to be the one inside the newest version folder (second screenshot), so the path to the running executable seems to change with every update. However, I’m not entirely sure about this, because there is also a Figma.exe in the root folder (C:\Users\<username>\AppData\Local\Figma\Figma.exe, screenshot below). Why this is a problemThird-party apps that identify Figma by its executable path stop working after an update. Initially, the third-party app targets  a Figma.exe in the root folder (C:\Users\<username>\AppData\Local\Figma\Figma.exe, screenshot above).

When using the new text, Several functions in design are not available😅

Opening the Windows Emoji Panel (Win+.) works, but clicking any emoji fails to insert it anywhere in Figma.Test and failed: Canvas text layers, layer names, page names, component names... (everywhere) Additional notes: This issue only occurs in the Figma Desktop App, browser works properly.  Figma desktop app Windows 11

I would like to volunteer for the Figma Config event and assist with event management. I’m happy to assist with crowd management, stalls, meals, or any other tasks where support is needed.  I’m not looking for compensation, I just want the opportunity to contribute and be part of Config India.  I would really appreciate the opportunity.

Hello Figma Community Support,The Figma Help Center assistant directed me here and stated that Community Support specialists can route this case to the plugin review team. Please help forward my request for clarification and reconsideration.Plugin: [redacted by moderator to remove personal information]Plugin ID: [redacted by moderator to remove personal information]Resource: [redacted by moderator to remove personal information]The rejection message states:“We're not able to approve apps that overlap with native capabilities in the Figma MCP server. Consider building skills on top of the official use_figma tool.”I understand the concern about overlapping functionality. However, this plugin was developed to address a local-font loading difference observed in testing on October 11, 2026:1. An existing native text node used the font family “HYDiShengYingXiongTi”, style “J”.2. The official use_figma workflow could read the node, but loading its original font with figma.loadFontAsync failed

Hi! Has the canvas panning keyboard shortcuts changed recently? I use Ctrl+Space+scroll wheel to pan and zoom the canvas but now that keyboard shortcut puts me into a Keyboard picking mode. Is this new? It really gets in the way of working.

Hi, I'm following up on my support request: https://help.figma.com/hc/en-us/requests/2170914I haven't received any updates so far, and the issue is still unresolved. Could you please look into this and provide an update as soon as possible?Thank you for your help!