Guide · Figma plugins

How do AI search tools find Figma plugins? The Community plugin page's robots.txt, sitemap and Figma's requirements

Updated 2026-11-13 · kagentx

Short answer

Tested on 2026-11-13: Figma's robots.txt leaves plugin pages open to ordinary crawlers, but the group shared by CHATGPT-User, OAI-SearchBot, PerplexityBot, Claude-SearchBot and Claude-User allows only the home page, blog, help and developer docs and ends with Disallow: /; GPTBot, ClaudeBot and Google-Extended each have Disallow: /. The Community plugin sitemap lists 18,445 URLs. We could not fetch a plugin detail page (HTTP 403), so its HTML is not verified. Put the key information on your own site; the Community page is only an entry point.

Is Figma open to crawlers?

Open to ordinary crawlers; for the main AI search crawlers it is closed, going by the file's text. On 2026-11-13 we read figma.com/robots.txt with curl (User-Agent Mozilla/5.0, 200, 394,916 bytes, 26 User-Agent groups):

CrawlerWhat robots.txt says
* (all crawlers not named)Blocks /exit?*, the Community search page /community/search?* (in every language), /api/*, /board/*, /proto/, /make/*, /file/new?try-plugin-id=* and more; does not block /community/plugin
GPTBot, ClaudeBot, CCBot, Google-Extended, cohere-ai, FacebookBot, Bytespider and othersOne group each, Disallow: /
CHATGPT-User, OAI-SearchBot, PerplexityBot, Claude-SearchBot, Claude-UserOne shared group: about 9,000 Allow lines (home, blog, help, legal pages, /developers/*, /plugin-docs/* and so on), ending with Disallow: /; the Allow lines in this group have no /community/plugin path

By the text of robots.txt, these five crawlers should not read plugin pages. This is inferred from the text and was not verified with real crawlers; what each crawler is for is in AI crawler user agents: a side-by-side table, how to write the rules is in How do you write robots.txt for AI crawlers?.

Does Figma have a plugin sitemap?

Yes. robots.txt lists three Sitemap lines; figma-community-index.xml is the Community index, and on 2026-11-13 we got:

CheckTest result
www.figma.com/sitemap.xml200, 1,768 bytes, a sitemap index listing each language version, lastmod 2026-11-11
figma-community-index.xml200, 3,880 bytes, 24 child sitemaps: plugins, widgets, FigJam files, Figma files (7 parts), slides, sites, categories, tags and more
community-plugins-sitemap.xml200, about 9.3 MB, 18,445 URLs, all with lastmod, newest 2026-11-13; 17,797 are /community/plugin/<numeric ID>/<name>, 648 are language versions

A plugin page's URL is “numeric ID + name”, and the name part comes from the plugin name, so the plugin name is also part of the URL.

What is in the HTML of a plugin detail page?

We could not read it: not verified (blocked). On 2026-11-13 we fetched a plugin page, the same URL without the name, the Community home page and a Community file page with curl and WebFetch, and all returned HTTP 403 (CloudFront's “The request could not be satisfied”, 919 bytes), while figma.com's home page returned 200 from the same machine. We do not work around verification or blocking, so we say nothing about whether the plugin page has a meta description, canonical or structured data, or whether its body is rendered by script.

What the Community file pages look like is in another guide (tested 2026-11-10); see How do you get Notion templates, Figma Community files and Framer templates recommended by AI?. Whether crawlers run scripts is in Do AI crawlers render JavaScript?.

What does Figma require for plugin listings?

Figma Help Center's publishing guide and review guidelines (read in full on 2026-11-13; neither page shows an update date):

ItemOfficial statement
Submission requirementsSubmit from the macOS or Windows desktop app; turn on two-factor authentication before publishing; plans are Starter, Professional, Education, Organization or Enterprise
FieldName, tagline, description, category (such as Design tools, Software development), support contact, publisher (yourself, a team or an organization); the page gives no length limits, does not mention tags, and does not say which fields are required
ImagesIcon recommended 128 × 128 px; cover (image or video) recommended 1920 × 1080 px; up to 9 images and videos in the carousel; the page gives no format or size limits
Data security and network accessThe data security form is reviewed by Figma and may take up to 2 weeks; the network access label on the listing is based on manifest.json, and if it shows “Unknown network access” you should declare network access in the manifest
Review processAfter submitting it shows “In review”, the result goes to your account email, timing varies; a rejected plugin can be revised and resubmitted; you can push updates during review; plugins shared privately inside an organization are not reviewed
Listing contentMust match what the plugin actually does; link to setup and usage documentation; disclose any third-party account or extra authorization needed in the description; provide a way to contact support; no temporary content; duplicate plugins share one listing
Legal and privacyFollow Figma's trademark guidelines when using the Figma brand; plugins that handle user data must provide and maintain a privacy policy

The vendor does not say how these fields affect AI citation, and we have no evidence: not verified.

Besides the Community plugin page, what other page does an indie developer need?

One full introduction page on your own site. The Community page is closed to the main AI search crawlers in robots.txt, and we could not read the page's HTML, so do not keep the key information only on the Community page.

  1. 1. Make the first sentence on your site a definition

    One sentence saying what it does and where in Figma it is used, matching the Community listing's tagline in meaning but not copied word for word.

  2. 2. Cover what the review guidelines require

    How to install and use it, any third-party account or authorization needed, what user data it handles, contact details, privacy policy. The review guidelines require the listing to link to these documents; keeping them on your own site is the safest. How to write documentation is in How should developer docs be written so AI can understand and cite them?.

  3. 3. Link your site and the Community page to each other

    Link from your site to the Community page, and put your site's URL in the Community page's description and support contact.

  4. 4. Add structured data and render the body on the server

    Describe the same plugin on the page with SoftwareApplication, written as in Should a SaaS add SoftwareApplication structured data?; do not rely on scripts to render the body.

  5. 5. Check that your site does not block AI crawlers

    Use the AI crawler checker on your own site's robots.txt; do not copy figma.com's rules.

  6. 6. Test again after a few weeks with the same questions

    Ask ChatGPT and Perplexity “is there a Figma plugin that does X”, 3 times per question, and note whether you are mentioned and which URLs are cited.

Compare with similar stores: How do AI search tools find JetBrains plugins? The Marketplace plugin page, tested, How do VS Code extensions get recommended by AI?, How do I get my Obsidian plugin found in AI search?.

If you build a Canva app, see How do AI search tools find Canva apps? The Apps Marketplace tested, plus the official requirements for the Apps Marketplace robots.txt and sitemap test and the official listing requirements.

FAQ

Can AI crawlers read a Figma Community plugin page?

Going by the text of robots.txt, the main AI search and user-agent crawlers cannot. On 2026-11-13 we tested figma.com/robots.txt: CHATGPT-User, OAI-SearchBot, PerplexityBot, Claude-SearchBot and Claude-User share one group that allows the home page, blog, help and developer docs and so on and ends with Disallow: /, with no /community/plugin path allowed; GPTBot, ClaudeBot, Google-Extended and others each have a Disallow: / group. This is inferred from the text and was not verified with real crawlers.

Does Figma's robots.txt leave plugin pages open to ordinary search crawlers?

Yes. The * group does not block /community/plugin paths; it blocks only the Community search page (/community/search?*), /api/*, /board/*, /proto/, /make/* and so on. Figma also lists three sitemaps, and the Community index contains a plugin sitemap. Whether Google's and Bing's crawlers can read the plugin page body is not verified.

Does a plugin page's HTML have a title, description and structured data?

Not verified (blocked). On 2026-11-13 we fetched a plugin page and Community pages with curl and WebFetch and got HTTP 403 (CloudFront), while figma.com's home page returned 200. We do not work around the block, so we did not read the plugin page's HTML.

Does Figma have a plugin sitemap?

Yes. The Sitemap line in robots.txt points to figma-community-index.xml, which lists 24 child sitemaps; community-plugins-sitemap.xml has 18,445 URLs, each with a lastmod, the newest being 2026-11-13.

Which fields do you fill in to list on Figma Community?

The vendor's page (read in full on 2026-11-13) lists: name, tagline, description, category, support contact, publisher; icon recommended 128 × 128 px, cover recommended 1920 × 1080 px, up to 9 images and videos in the carousel. The page gives no length limits and does not say which fields are required.

How long does review take?

The vendor's page says that after submitting the plugin shows “In review”, the result goes to your account email, and timing varies; the data security form review may take up to 2 weeks. Plugins shared privately inside an organization are not reviewed.

Besides the store page, what other page does an indie developer need?

One full introduction page on your own site: what it does, how to install it, which permissions it needs, changelog, contact, privacy policy. The review guidelines require the listing to link to setup and usage documentation, name any third-party account needed, and provide a privacy policy if user data is handled; keeping these on your own site is the safest.

Sources

kagentx

Can crawlers read your pages?
Let the evaluation tell you

The evaluation checks whether crawlers can get in and whether the text is served directly, then asks ChatGPT, Perplexity and Claude (with web search on) your buyers' questions and sees whether you are mentioned. The Basic plan is $1, and results usually take 2–3 minutes.