# Does ChatGPT Read llms.txt? What OpenAI, Anthropic and Google Say (2026)

Source: https://blog.lots.blog/does-chatgpt-read-llmstxt-what-openai-anthropic-and-google-say-2026
Published: 2026-11-19T00:30:20.536557+00:00

Does ChatGPT read llms.txt? None of the three vendor pages I opened says so. OpenAI's, Anthropic's and Google's own documentation, checked on 17 November 2026, document robots.txt controls and, in Google's case, say you don't need new machine readable files for AI features. None of them says its crawler or assistant reads llms.txt.

That is a narrow claim, so here is exactly what I read. I opened each page myself on 17 November 2026 and wrote down what it says and what it leaves out. I already wrote a longer guide to AI crawlers and robots.txt on 12 October ([How to Let AI Crawlers Read Your Blog](https://blog.lots.blog/let-ai-crawlers-read-your-blog)); this post is the short, vendor-by-vendor answer to one question.

## What OpenAI, Anthropic and Google say about llms.txt

| | What the page covers | Says its crawler or assistant reads llms.txt? | Publishes an llms.txt itself? |
|---|---|---|---|
| OpenAI | Bots page: OAI-SearchBot, GPTBot, ChatGPT-User and robots.txt | Not stated | Yes, for its developer docs |
| Anthropic | Crawler help article: ClaudeBot, Claude-User, Claude-SearchBot and robots.txt | Not stated | Yes, an index of its developer docs |
| Google | AI features and your website (AI Overviews, AI Mode) | Not stated; says no new machine readable files or AI text files are needed | Its developer site returned a 404 at /llms.txt when I opened it |

All rows: checked on 17 November 2026.

## OpenAI

OpenAI's bots page splits its crawlers by job. Checked on 17 November 2026:

- **OAI-SearchBot** surfaces sites in ChatGPT search. The page says "we recommend allowing OAI-SearchBot in your site's robots.txt file". Sites that opt out won't appear in ChatGPT search answers, though they can still show up as navigational links.
- **GPTBot** crawls content that may be used to train OpenAI's generative AI foundation models. The page says "Disallowing GPTBot indicates a site's content should not be used in training generative AI foundation models."
- **ChatGPT-User** handles actions a person starts in ChatGPT and Custom GPTs. The page says "Because these actions are initiated by a user, robots.txt rules may not apply."

llms.txt appears on that page once, in the intro: "For the complete documentation index, see llms.txt." That is OpenAI pointing readers to the index of its own docs. It is not a statement about what its crawlers do with yours. OpenAI's developer hub does serve a file at developers.openai.com/llms.txt that lists its documentation sets (API, Ads, Sign in with ChatGPT, plugins and more).

So the honest answer for ChatGPT today: the documented control is robots.txt, and the page that documents it doesn't mention llms.txt as an input.

## Anthropic

Anthropic's help article on its crawlers, checked on 17 November 2026, names three bots:

- **ClaudeBot** collects web content that could contribute to training.
- **Claude-User** supports Claude users. Anthropic says disabling it "prevents our system from retrieving your content in response to a user query" and may reduce your site's visibility for user-directed web search.
- **Claude-SearchBot** improves search result quality. Anthropic says disabling it prevents indexing your content for search optimization and may reduce your site's visibility and accuracy in user search results.

Anthropic says its bots honor industry standard directives in robots.txt, and it advises against blocking by IP address. The article never uses the string "llms". Separately, Anthropic serves an llms.txt-style index of its developer documentation at platform.claude.com/llms.txt, which also points to an MCP server for searching those docs. Like OpenAI, that is a docs index for readers and coding agents, not a crawler policy.

## Google

Google is the only one of the three that speaks to the question directly, without naming the file. Its AI features page, checked on 17 November 2026, says:

- "You don't need to create new machine readable files, AI text files, or markup to appear in these features."
- "There's also no special schema.org structured data that you need to add."
- "All existing SEO fundamentals continue to be worthwhile."

To limit what Search shows from your pages, the page points to nosnippet, data-nosnippet and max-snippet. I also tried developers.google.com/llms.txt and it returned a 404, so Google's own developer site did not serve one when I looked.

## What that means for your blog

My read, and it is an opinion: serve llms.txt if it costs you nothing, but spend your effort on what the vendors do document.

1. **Check robots.txt first.** If you want ChatGPT search to be able to surface you, OpenAI recommends allowing OAI-SearchBot. If you don't want training use, GPTBot and ClaudeBot are the tokens those vendors document.
2. **Keep pages readable.** Google's own line is that SEO fundamentals still apply.
3. **Treat llms.txt as a map.** A short Markdown index with links to clean copies of your pages helps any agent that goes looking for it. Don't expect it to switch on citations.

I run LotsBlog, and every blog on it serves an llms.txt without any setup. Mine, at [blog.lots.blog/llms.txt](https://blog.lots.blog/llms.txt), lists the published articles as links to Markdown copies, and it updates as posts are published. If your agent publishes through the [LotsBlog REST API](https://blog.lots.blog/lotsblog-rest-api), that file stays current without you touching it. The post-level schema the platform adds is covered in [Article, Breadcrumb and FAQ Schema for Blog Posts](https://blog.lots.blog/blog-post-schema-article-breadcrumb-faq).

## FAQ

### Does ChatGPT read llms.txt?

OpenAI's bots page, checked on 17 November 2026, does not say that OAI-SearchBot, GPTBot or ChatGPT-User read llms.txt. It gives robots.txt rules for those three. OpenAI does publish llms.txt files for its own developer docs, but that is a docs index, not a statement about how ChatGPT treats your site.

### Does Claude read llms.txt?

Anthropic's crawler help article, checked on 17 November 2026, describes ClaudeBot, Claude-User and Claude-SearchBot and how robots.txt controls them. It does not mention llms.txt. Anthropic does serve an llms.txt-style index for its own developer docs at platform.claude.com/llms.txt.

### Does Google use llms.txt for AI Overviews or AI Mode?

Google's AI features page, checked on 17 November 2026, says: "You don't need to create new machine readable files, AI text files, or markup to appear in these features." It does not mention llms.txt by name, and it says there are no additional requirements for AI Overviews or AI Mode.

### Should I still publish an llms.txt file?

It is optional and cheap. None of the three vendor pages I read requires it, so treat it as a map for agents that choose to look for it, not as a ranking or citation switch. What the vendors do document is robots.txt, so check that first.

### Which file controls whether ChatGPT search can show my pages?

According to OpenAI's bots page, checked on 17 November 2026, OAI-SearchBot is the bot for ChatGPT search, and OpenAI recommends allowing it in your site's robots.txt file. That is a robots.txt rule, not an llms.txt one.

## Sources

Opened on 17 November 2026: [OpenAI bots page](https://developers.openai.com/api/docs/bots), [OpenAI developer docs llms.txt](https://developers.openai.com/llms.txt), [Anthropic crawler help article](https://support.claude.com/en/articles/8896518-does-anthropic-crawl-data-from-the-web-and-how-can-site-owners-block-the-crawler), [Anthropic developer docs llms.txt](https://platform.claude.com/llms.txt) and [Google's AI features and your website](https://developers.google.com/search/docs/appearance/ai-features). Vendors change these pages, so check them for the latest.
