To get indexed by ChatGPT, let OpenAI's search crawler, OAI-SearchBot, read your site through robots.txt, and make sure your pages show their text in plain HTML. Then verify your site in Bing Webmaster Tools and submit your sitemap. That is how to get indexed by ChatGPT. Everything else only helps your pages get picked once they are in the index.
There is no form to fill in and no "ChatGPT Search Console". The work is about removing blocks: a robots.txt rule, a firewall setting, a security plugin, or a page that only loads with JavaScript. On a normal business site, the checklist below takes about 30 minutes. It also shows which steps come from OpenAI's own documentation and which are educated guesses.
Indexed, cited, recommended: three different things
People mix these up, and each one needs different work.
- Indexed: OpenAI has crawled your page and stored it, so ChatGPT search can find it. HubSpot makes the same split between being stored in the index and being quoted in an answer.
- Cited: ChatGPT uses your page as a source in an answer and links to it.
- Recommended: ChatGPT names your business when someone asks for a provider, a tool or a shop.
This guide covers step one, getting into the index. Nothing else can happen until that works. Once you are in, citations depend on how useful your pages are. Our guide on how to get cited by ChatGPT covers that part. If you would rather have someone else write pages built to be quoted, see how LLM SEO with AI SEO Crew works.
The three OpenAI crawlers and what each one does
OpenAI documents three user agents on its crawlers page. You set rules for each one separately.
| Crawler | What it does | Affects ChatGPT search? | Follows robots.txt? |
|---|---|---|---|
| OAI-SearchBot | Finds pages to show in ChatGPT's search features | Yes, this is the one that matters | Yes |
| GPTBot | Collects content that may be used to train OpenAI's models | No | Yes |
| ChatGPT-User | Visits a page when a user asks ChatGPT to open it | Only for that one request | Not always: OpenAI says robots.txt rules may not apply because a user starts the visit |
The key line from OpenAI: a site that opts out of OAI-SearchBot will not appear in ChatGPT search answers, though it may still show up as a plain navigational link (source). OpenAI also says allowing OAI-SearchBot is not strictly required, but that it helps your visibility. In practice, if you want to show up, let it in.
The robots.txt to copy (search yes, training optional)
Search and training are set separately. You can be found in ChatGPT search without letting OpenAI train on your content. This robots.txt block does exactly that:
User-agent: OAI-SearchBot
Allow: /
User-agent: ChatGPT-User
Allow: /
User-agent: GPTBot
Disallow: /
Sitemap: https://www.yoursite.com/sitemap.xml
If you don't mind training, change the GPTBot line to Allow: /. Either way, it has no effect on search.
While you are in the file, check three things:
- Look for a global block. A line like
User-agent: *followed byDisallow: /shuts out every crawler that has no group of its own. A named OAI-SearchBot group, as above, gives it its own rules. - Keep your normal rules. If you block
/cart/or/admin/for everyone, copy those lines into the OAI-SearchBot group too. - Wait a day. OpenAI says a robots.txt change can take about 24 hours to reach its search systems.
Where to edit robots.txt on your platform
Every builder puts the file in a different place, and menus change often. The rule of thumb: open your platform's SEO or site settings and look for a robots.txt field. If you can't find one, search your platform's help center for "robots.txt".
| Platform | Where to start looking | What to watch |
|---|---|---|
| WordPress | WordPress serves a virtual file. Most SEO plugins have a robots.txt editor, or you can upload a real robots.txt to the site root |
Security and firewall plugins can block bots even when robots.txt allows them |
| Webflow | The SEO part of your site settings | Make sure indexing is not turned off for your live domain |
| Wix | The SEO settings of your site | Check that each important page is not hidden from search engines |
| Framer | Generated for you. A custom file can be uploaded through Static Files on Pro and Enterprise plans (Framer help) | Your upload replaces the generated file, so keep the Sitemap line |
| Shopify | Your theme settings or theme code, depending on how your store is set up | Keep Shopify's default rules when you add your own |
| GitHub sites (Next.js, Astro, Hugo, Jekyll) | A robots.txt file in the public or static folder, or a generated route |
Rebuild and check the live file, not the one in your repo |
When you are done, open yoursite.com/robots.txt in a browser and read what is actually live.
Bing Webmaster Tools: verify and submit your sitemap
OpenAI's crawler page does not mention Bing. Rank Math's ChatGPT indexing guide still lists Bing indexing as a requirement, and it costs little either way. Treat it as strongly recommended, not officially required.
- Sign in to Bing Webmaster Tools and add your site. If you already use Google Search Console, you can import your site from there instead of verifying again.
- Submit your sitemap URL in the Sitemaps section.
- Use URL inspection on your most important page to check that Bing can fetch it.
- Come back after a few days to see how many pages are indexed and which ones were excluded, with the reason.
If Bing excludes pages, fix those first. A page Bing refuses to index usually has a problem other crawlers will hit too. Our guide on blog posts not indexed by Google covers the usual causes.
While your sitemap is fresh in mind, you can also list your key pages for language models with our free llms.txt generator. It takes a few minutes. It won't get you indexed on its own (more on that below), but it does no harm. To check what crawlers actually see on a page, run the free SEO checker: it shows whether robots.txt allows OAI-SearchBot, Googlebot, Bingbot and PerplexityBot, and how much readable text your HTML contains.
Check that your pages work without JavaScript
HubSpot reports that OpenAI's crawlers don't run JavaScript. If your text only appears after a script runs, the crawler sees an empty page.
A two-minute test anyone can do:
- Open a blog post and press Ctrl+U (Cmd+Option+U on a Mac) to view the page source.
- Press Ctrl+F and search for a sentence from the middle of the article.
- If you find it, the text is in the raw HTML. If you don't, the page relies on JavaScript.
WordPress, Webflow, Wix and Framer pages usually pass. The risk is higher with single-page apps, custom React sites and widgets that load content after the page opens. On a Next.js or Astro site, make sure posts are rendered on the server or built as static pages.
Find out if a CDN, firewall or plugin is blocking OAI-SearchBot
Many sites allow OAI-SearchBot in robots.txt and still block it at the door. robots.txt is a polite request. Firewalls don't read it; they decide on their own.
- Cloudflare: AI Crawl Control shows which AI crawlers visit your site and lets you allow or block each one. Open it, find OAI-SearchBot and make sure it is allowed. If you turned on any broader bot blocking, check that it does not catch OAI-SearchBot.
- WordPress security plugins: look in the firewall or bot rules for blocked user agents and rate limits that catch crawlers.
- Hosting firewalls: some hosts block unknown bots by default. Your host's support team can tell you whether OAI-SearchBot gets through.
How to confirm ChatGPT can find you
There is no OpenAI dashboard to look things up in, so you check the signals instead.
- Server logs. Search your access logs (or your host's bot report) for
OAI-SearchBot. Visits with status 200 mean it can read your pages. A pile of 403 or 429 codes means something is blocking it. - IP ranges. Anyone can fake a user agent. To confirm a visit is real, compare its IP with the list OpenAI publishes at openai.com/searchbot.json. Lists for the other crawlers sit at
openai.com/gptbot.jsonandopenai.com/chatgpt-user.json(source). - Referral traffic. In Google Analytics, look for sessions where the source is chatgpt.com. Any visits at all prove your pages are being shown.
- Real questions. Ask ChatGPT, with search on, the questions your pages answer, in your customers' words. Check the sources it lists. Try a few phrasings and repeat every week, because answers change.
How long does it take to get indexed by ChatGPT?
The parts you control are fast. A robots.txt change reaches OpenAI in about 24 hours (OpenAI). Being cited takes longer. HubSpot reports that a page can be indexed within hours, but in the study it cites, the median time to a first citation was 6.81 days after publishing (HubSpot). That is one third-party study, not a promise. Some pages are cited sooner, some much later, and some never.
Do you need to submit your site to ChatGPT?
No. OpenAI has no submission form and no webmaster console for ChatGPT search. Your sitemap does the job of a submission: list it in robots.txt and submit it to Bing. Anyone selling "ChatGPT submission" is selling something OpenAI does not offer.
What is official and what is a guess
| Tactic | Status |
|---|---|
| Allow OAI-SearchBot in robots.txt | Official, from OpenAI's crawler docs |
| Block GPTBot while allowing search | Official, the two are set separately |
| Bing indexing and sitemap | Widely recommended, not mentioned on OpenAI's crawler page |
| Text visible without JavaScript | Reported by HubSpot, and easy to test yourself |
| llms.txt file | No official support from OpenAI |
| Backlinks and brand mentions | Likely helpful for citations, not for indexing itself |
If you want the details on llms.txt, read does llms.txt work.
Once you are indexed, the content does the rest. OpenAI has not published how ChatGPT picks its sources. The habits our crew writes by are a best practice, not a rule from OpenAI: answer the question in the first two sentences, give short definitions, link each fact to its source, use clear headings and lists, and keep pages up to date. The same habits make pages easier to use for Perplexity and Google's AI Overviews.
How AI SEO Crew handles this
Indexing is step one. Pages worth quoting are step two, and that is the part the crew writes. Each article opens with a direct answer, links its facts to sources and uses a clear structure. The Editor checks every fact and link before anything goes live. The Publisher then publishes to your site, sends your sitemap to Google and pings IndexNow, so crawlers hear about new pages quickly. Nobody can guarantee a citation in ChatGPT. You still own your robots.txt and firewall settings, so run the checklist above once.
Start with robots.txt, Bing and the firewall check today, then publish pages worth quoting. Start free and see your SEO plan in a few minutes, no card needed.
Frequently asked questions
Is there a ChatGPT Search Console?
No. OpenAI offers no dashboard that shows which pages are indexed. Use your server logs, OpenAI's published IP lists, chatgpt.com referrals in your analytics, and real test questions instead.
Can I block ChatGPT from training on my pages but still be cited?
Yes. Allow OAI-SearchBot and disallow GPTBot in robots.txt. OpenAI documents them as separate crawlers, and blocking GPTBot does not remove you from ChatGPT search.
Does llms.txt help ChatGPT index my site?
There is no official evidence that it does. OpenAI's crawler documentation is built around robots.txt, not llms.txt. Treat it as an optional extra once robots.txt, Bing and plain HTML are in place.
Do backlinks still matter for ChatGPT indexing?
Not for indexing itself, which depends on crawl access. Links and mentions on other sites may help your pages get chosen as sources, but OpenAI has not published how it ranks or picks sources.