Home / Guides

How to show up in ChatGPT search results: 7 steps

ChatGPT can search the web and cite pages in its answers. Whether your site can be one of those pages starts with two settings: one line in robots.txt and one rule in your firewall.

This guide covers what OpenAI's own crawler docs say, the seven settings to check, and what we found when we scanned 500 software company homepages on October 18, 2026. No one can promise a citation. These steps remove the reasons ChatGPT can't use your pages.

What OpenAI says decides it

OpenAI runs a crawler called OAI-SearchBot for search. Its >crawler docs say:

"OAI-SearchBot is used to surface websites in search results in ChatGPT's search features. Sites that are opted out of OAI-SearchBot will not be shown in ChatGPT search answers, though can still appear as navigational links."

The same page recommends "allowing OAI-SearchBot in your site's robots.txt file and allowing requests from our published IP ranges." Those two settings are the base. The rest of this guide helps any crawler read and describe your pages.

The 7 steps

1. Allow OAI-SearchBot in robots.txt.
2. Let OpenAI's published IP ranges through your firewall or CDN.
3. Decide on GPTBot (training) as a separate choice.
4. Know what ChatGPT-User does and doesn't control.
5. Put your main text in the HTML, not in JavaScript alone.
6. Give crawlers a sitemap, a meta description, and structured data.
7. Wait about 24 hours and check your server logs.

Step 1: allow OAI-SearchBot in robots.txt

Open yoursite.com/robots.txt. If you see User-agent: OAI-SearchBot followed by Disallow: /, ChatGPT search can't show your pages in answers. A blanket rule can do the same thing:

User-agent: *
Disallow: /

That blocks every crawler that obeys robots.txt, OAI-SearchBot included. To allow it in plain terms, add a group of its own:

User-agent: OAI-SearchBot
Allow: /

Rules apply to the User-agent lines right above them, so a bot with its own group follows that group. Check your file with the AI crawler robots.txt checker, and see the OAI-SearchBot page for the user agent details.

Step 2: let OpenAI's IP ranges through your firewall

A robots.txt rule does nothing if your firewall blocks the bot before it reaches the page. OpenAI publishes the IP ranges for OAI-SearchBot at >openai.com/searchbot.json and recommends allowing requests from them.

If you use Cloudflare, Vercel, Akamai, or another bot shield, look for an "AI bot" or "bot fight" setting and check what it does to verified search crawlers. In our scan, 20 of 500 homepages showed our checker a bot challenge page in place of the site: 16 from Cloudflare, 2 from Vercel, 1 from Akamai, and 1 we could not name. Our checker is not OAI-SearchBot, so this does not prove OpenAI was blocked. It does show how common these walls are.

Step 3: treat GPTBot as a separate choice

GPTBot collects pages that may train OpenAI's models. OAI-SearchBot is for search. OpenAI says "each setting is independent of the others," so you can block training and keep search:

User-agent: GPTBot
Disallow: /

User-agent: OAI-SearchBot
Allow: /

In our scan, 7 of 467 scored sites blocked GPTBot: Loom, Moosend, Vidyard, Figma, Kolide, Bardeen, and June. All 7 left OAI-SearchBot allowed. Our AI crawlers in robots.txt guide has rules for the other AI companies.

Step 4: know what ChatGPT-User does

ChatGPT-User is the agent that visits a page when a person asks ChatGPT about it. OpenAI says it "is not used to determine whether content may appear in Search" and that, because a person starts these visits, "robots.txt rules may not apply." Blocking ChatGPT-User does not keep you out of search, and allowing it does not put you in. Two of the 467 sites in our scan blocked it: Moosend and June.

Step 5: put your main text in the HTML

Some sites send an empty page and build the text with JavaScript in the browser. A crawler that does not run JavaScript sees little or no text. OpenAI's docs do not say whether OAI-SearchBot runs JavaScript, so the safe choice is to serve your main text, headings, and prices in the HTML.

Our checker tests this on every scan. 10 of 467 scored homepages had little or no text without JavaScript, among them PostHog, Bubble, and OneLogin. To test your own page, view the page source (not the inspector) and search for a sentence from your homepage.

Step 6: give crawlers a map and a summary

OpenAI's crawler docs do not list a sitemap, a meta description, or structured data as requirements. They are the standard ways to help any search crawler find your pages and describe them:

• Sitemap. List your pages in sitemap.xml and point to it from robots.txt with a Sitemap: line. 23 of 467 sites had none we could find.
• Meta description. One or two plain sentences on what the page offers. 10 of 467 homepages had none.
• JSON-LD structured data. Mark up your organization, products, and FAQs. 101 of 467 homepages (22%) had no JSON-LD.

351 of 467 sites (75%) passed all five of these checks: OAI-SearchBot allowed, text in the HTML, a sitemap, a meta description, and JSON-LD.

Step 7: wait and check your logs

OpenAI says it "can take ~24 hours from a site's robots.txt update for our systems to adjust." After a day, search your server or CDN logs for OAI-SearchBot. Its user agent ends in compatible; OAI-SearchBot/1.4; +https://openai.com/searchbot (OpenAI notes the version number may change), and requests for robots.txt can carry an extra robots.txt marker. Anyone can fake a user agent, so match the IP address against searchbot.json.

If the bot reaches your pages and gets a 200 response, the access side is done. What ChatGPT picks to cite after that depends on its own systems, which OpenAI does not publish.

Does llms.txt help with ChatGPT search?

OpenAI's crawler docs do not mention llms.txt. It is a proposed file that gives AI tools a short map of your site. In our scan, 261 of 467 sites (56%) had one. It costs little to add, but it does not replace the steps above. See llms.txt example and format if you want one.

What 467 software sites get right and wrong

From the AI readiness leaderboard scan on October 18, 2026 (500 homepages tried, 467 scored):

Robots.txt is not the problem for these sites. The common gaps are missing llms.txt and JSON-LD, plus the 20 homepages that put a bot challenge in front of our checker. The full data is free to reuse under CC BY 4.0 at leaderboard.json.

Check your site

Run the free AI readiness check on your domain. It tests OAI-SearchBot and seven other AI user agents in robots.txt, text without JavaScript, JSON-LD, sitemap, meta description, and llms.txt, and tells you when a challenge page blocks it. The deep report crawls 25 pages and lists fixes in order.

FAQ

How do I get my website to show up in ChatGPT search? Allow OAI-SearchBot in robots.txt and let OpenAI's published IP ranges through your firewall. OpenAI says sites that opt out of OAI-SearchBot will not be shown in ChatGPT search answers.

Does blocking GPTBot remove me from ChatGPT search? No. OpenAI says GPTBot and OAI-SearchBot are separate settings. You can block GPTBot for training and allow OAI-SearchBot for search.

How long does a robots.txt change take to work in ChatGPT search? OpenAI says about 24 hours.

Can I pay to appear in ChatGPT search results? OpenAI's crawler docs describe no paid inclusion for search answers. They describe OAI-AdsBot, which checks pages submitted as ads on ChatGPT. That is a separate ad product.

Can ChatGPT link to my site if I block OAI-SearchBot? OpenAI says opted-out sites "can still appear as navigational links." Your content will not be used in its search answers.

Checked 2026-10-18. Prices and policies change. Confirm on the source before you buy.

Sources

• https://developers.openai.com/api/docs/bots
• https://openai.com/searchbot.json
• https://checkaiready.com/leaderboard.json

More guides

• AI crawlers in robots.txt: GPTBot, ClaudeBot, PerplexityBot rules (2026)
• llms.txt example and format: how to write one that passes (2026)