Share Your Thoughts

Feature Request to Resolve Ongoing Library Problem: Allow Swap Library to Replace Local Variables with Published Library Variables I would like Swap library to support replacing local variables in a file with variables from a published library when the collection structure and variable names match.
Today, Swap library only supports library-to-library swaps, which means local variables cannot be migrated directly to an existing published library.
This creates extra overhead because the workaround is to publish the local variables as a separate library first, only to swap them afterward.
 Since variable replacement by matching names already works within a local file, extending that behavior to local-to-library swaps would make migration to standardized library variables much easier. Example:A file has local variables in a collection called “Module 1 Semantic Theme.” That file also has layers using local variables like “Color / Text / Primary” and “Color / Background / Default.” Your publ

When working with AI Agents, I often return to the same chat over a longer period of time. As more chats are created, it becomes difficult to quickly find the conversation I’m actively working on, especially when switching between Agents and Files.It would be helpful to:Pin important chats so they stay at the top for quick access. Rename chats so ongoing conversations can be clearly identified by project or task.These options would make it much easier to organize and return to ongoing AI Agent work without having to search for the same chat each time.

When editing text in a Motion scene, clicking a specific text element often selects another text layer that is higher in the stack.To edit the intended text, I currently have to hide the other scenes/layers first, make the change, and then show them again. This becomes time-consuming when working with multiple scenes.It would be helpful if we could directly select and edit the text element we click without having to hide other scenes or layers. 

Am I the only one who’s left wanting by the new position sticky feature? In my opinion, there are three main problems with the new sticky feature: It’s not compatible with auto layout SolutionMake the sticky elements always appear on top of other elements (and later add a z-index feature so we can fine tune the behavior) The sticky element has to be an immediate child of the scrolling container SolutionMake the sticky behavior relative to the closest scroll container instead, so the sticky element can be a grand child, great grand child etc. Only works for vertical scrolling and can only stick to the top of the container SolutionAdd an offset property (top/right/bottom/left) - like the CSS implementation - super simple, but pretty flexible (only problem with the CSS implementation is the overflow: hidden limitation) What I want to be able to make None of which is possible with the new sticky feature unfortunately HTML prototype Note the sticky footer inside th

Add richer descriptions to variable collections, groups, and modes Figma currently allows us to add a description and scope to an individual variable. However, the description field is a relatively limited text input, and documentation cannot be added to the broader structures that give variables their meaning.Proposed Description LevelsIt would be helpful to support descriptions at multiple levels of the variable system: Variable collections: Explain the collection’s purpose, intended use, architecture, and relationship to other collections. Variable groups: Document how a specific category of variables should be applied and distinguish it from related groups. Modes: Explain the context represented by each mode, when it should be applied, and any important differences or limitations. Individual variables: Retain variable-specific guidance, but provide a more capable description experience comparable to component documentation. Ideally, these descriptions would support multiline

I’ve just applied variables to all our project’s text styles. It all works as expected. However, in inspect mode, font weight it still displayed as normal css value and not with css-variable. See screenshots. Is this a bug or me messing up things?

We are successful in accessing the figma remote MCP exposed at "https://mcp.figma.com/mcp" via OAuth based access via browser SSO login. We are exploring ways to automate figma MCP access in a headless mode so would like to understand passing PAT token for remote MCP authentication. We tried passing PAT in Auth headers and X-Figma-Token header but get unauthorized for remote MCP at this endpoint "https://mcp.figma.com/mcp". We are able to use the dev mode and local MCP with the figma desktop app as explained above we need to access these tools in a headless mode and would understand if there are any settings that needs to be enabled at the admin level to support PAT for remote MCP auth. ❯ curl --location 'https://mcp.figma.com/mcp' \--header 'Content-Type: application/json' \--header 'X-Figma-Token: figd_xxxx' \--header 'Accept: application/json' \--data '{    "jsonrpc": "2.0",    "method": "tools/list",    "params": {},    "id": "1"  }'Unauthorized❯ curl --location 'https://mcp.figma.

I have a feeling neither of the following are possible due to the current flexibility of using template slides -- ie you can choose to use just a few slides from a template instead of using them all. I’ve just published a large slides template for company wide use but realized it does not retain the nested sections or animations between slides. The deck has 91 slides so without the nested sections that appear under each title slide it’s difficult to manage.I also have some slides that use variants of the the same component that I have set to smart animate between then to show the proper animation, but this is not retained in the template either. Is it possible to keep either of these baked into the template? Thanks

It doesn’t look like there’s an obvious way to change the order of exposed nested instances. Right now the order seems to be based on reverse layer order. It’d be great if the order can be manually adjusted.

Hi everyone,I'm looking for a way to connect individual objects across different frames in a Design File to show UI flows, and I can't find a solution that ticks all these boxes:Machine-readable — the connection needs to exist as structured data queryable via the REST API / Plugin API (e.g. a node with startNodeId/endNodeId references), not just a visual line. Element-to-element across frames — connecting a specific object inside one frame to a specific object inside a different frame, not just frame-to-frame. Visible without Edit access — viewers with view-only access need to see the connections. Wrappable as a Component — for reusability with Component Properties (e.g. trigger, condition labels). Design Files only — not FigJam, since the connections need to live in the same file as the actual designs.What I've tried:Prototype connections — not visible without Edit access, and hard to query for additional logic. Native Connector tool — anchors to top-level frames but not to nested ele

I would love to see a native link-shortening feature built directly into Figma. Currently, we have to rely on third-party plugins or external tools to shorten share links. Having a built-in "Copy short link" option in the Share menu would make our workflow much smoother!

Auto-Layout has padding (inner spacing), but no margins (outer spacing). Currently the only 2 ways to add spacing is to: Create a spacer element or Wrap the items in another frame and apply gap spacing This creates a lot of unnecessary frames, and sometimes creates nesting so deep it goes outside the layers panel. Not sure if a technical limitation left this out of consideration in the first few versions of Auto Layout, but it would be much appreciated if it were included in a future version as it comes up as a significant limitation many times a day.

Create tags that can be added to post it notes for user research analysis

1. Describe the problem your experiencing and how your idea helps solve this If I add an image to figma and then resize that image I would like to export it at the original resolution. Maybe if there was even a button to resize the image to the full size. 2. Add as much context as possible (screenshots, Figma files, mockups, etc.) **3. Ask questions to bring the community into the conversation ** (e.g. Does anyone else experience this? Is anyone using a plugin or workaround to help solve this?, etc.

I’m working with a team on a large project, and we have a complex file structure. Removing the file previews has made it much harder to navigate our file system. Each group of files has its own color; previously, you could just glance at the main screen and understand where to go or see the file name in the preview. With only the folder icon and name, it’s much harder to remember where something is located. Please give us the option to revert to the previous interface with clear file previews.  

The UI for adding slides shows a grid of template options, but the previews are relatively small and no metadata is provided about the options (e.g. name). This makes it hard to identify the slide you want where there are many similar looking options or variants. I’d like to have either a large slide-out-on-hover preview or the name of the slide show on hover.  

Would love to see some added features for sharing Make designs with external members of your org without fully publishing it unprotected to the internet. Creating proofs of concept is much easier and faster in Make for some scenarios (for example– forms that react to the input) but are useless for using in testing because there are too many security concerns with publishing the link outside of the organization. Could it be only published for a certain pre-determined period of time? Could it be password protected with a password that auto-regenerates every so often? Could we have a one-time use link?Not all of us are using Make as a final product producer, but a fast way to create testable interactions that would be difficult or impossible with regular prototyping.

We are looking to setup an interactive Figma prototype of our new app so that we can quickly try out new ideas without code. We were very excited to learn about Figma Make via the upsell in the prototyping panel! Mainly because it would enable our coder colleagues not familiar with Figma to contribute on some design work. That could be such a productivity booster!But oh boy the disappointment when we learned that the purpose of Make is to generate application code instead of something that is useful for us to build upon. This is a weird direction for a design tool. How did you come to the conclusion that anyone would want to implement the application first and then manually create the designs and prototypes to iterate on? Make should output a plain Figma Prototype that we can build upon. AI can not replace the user testing cycle. Once happy with the design and ready to implement, it would indeed be very useful with a tool that generates the application skeleton from the prototype.

Now that our company has transitioned to Figma Enterprise across our entire design and development teams, we were eager to explore using Figma Sites to host a small internal dashboard and marketing toolkit website.However, we have run into a significant blocker: the native CMS does not support uploading PDFs or general document files.While using external cloud storage links (like Google Drive or AWS S3) works as a temporary band-aid, it is not a scalable or seamless workflow for a modern CMS platform. To truly compete as a viable website builder, Figma Sites needs the native ability to accept, host, and serve standard file types alongside images.This functionality is critical for numerous use cases:Internal Tools & Portals: Teams cannot build marketing hubs, asset dashboards, or resource toolkits without hosting downloadable deliverables (e.g., finalized PDFs, pitch decks, guides). Designer Portfolios: Freelancers and agencies using Figma to build portfolios desperately need a clea

I want to ask if you can please develop Figma more such that we can be able to draw a 2D like a circle in metric units for 1 to 1 scale. Also the modelling enviroment needs to be infinite for large 2D Drawings

Hi,We have thousands of icons as components in our library that should not be linked via Code Connect.It would be great to have a way to multi-select them and apply a batch action, such as “Skip connection”.At the moment, I would need to click on each icon, select “Skip connection” on the next screen, then go back to the list… Something like this, used in combination with filtering by name would be great!Thanks

The title case isn’t accurate. As it also capitalizes: articles (such as “a,” “an,” and “the”) conjunctions (such as “and,” “but,” and “or”), and prepositions (such as “at,” “by,” “of,” “to,” and “on”). In my opinion, the better naming would be “Capitalize”. Figma team can also consider adding “Title Case” and “Sentence Case” options separately.

Hello,I am looking to create a good hero section for my site, where it should take the height of browser in the first fold. The subsequent sections can be of fixed heights based on content. I need something like 100vh to have for my hero section.

Hi everyone! 👋The recent update allowing creators to showcase their design tool stack in Figma Community profiles («My Toolbox») is a fantastic step toward making profiles feel like true digital portfolios. However, the current selection is quite limited (28 preset icons) and primarily covers mainstream tools.Modern design workflows have evolved far beyond the traditional suite. Many designers today rely on a diverse, hybrid ecosystem that includes specialized LLMs, dedicated reference managers, AI prototyping tools, and knowledge management systems.To make «My Toolbox» truly representative of how creators work today, I’d love to propose expanding the selection to include a broader range of global tools, categorized below:1. Global LLMs & AI Assistants: DeepSeek (at the moment — V4-Flash-Vision), Qwen, Kimi, Meta AI, Grok, Agnes AI, Xiaomi MiMo, GLM, Microsoft Copilot. 2. AI-Powered UI, Code & Generation Tools: Cursor, Design Arena, Google Antigravity (2.0 & IDE), Leonardo

Yet another UI update that ruins the workspace.Now a whole column width is reducing the workspace size, created to accommodate one AI agent button that I have no interest in using. You’ve moved the library and assets to a less usable place  because you couldn’t figure out how to add the AI agent somewhere else? How can I roll back this change to the version that actually worked?