# Do I need an llms.txt file to be cited by ChatGPT or Claude?

No evidence shows you need one. llms.txt is a proposal published at llmstxt.org, the OpenAI and Anthropic crawler documentation does not list it as a signal, and Google says no special AI text files are needed for its AI features. It costs little to publish and helps agents that fetch your docs, but do not expect it to earn citations alone.

By SEOBolt. Published 13 November 2026, updated 13 November 2026.

## Why this happens
The /llms.txt idea came from Jeremy Howard in September 2024: a markdown file at the root of a site that gives a language model a short, curated map of the site and links to clean markdown versions of key pages. The proposal has been revised since; version 2 on llmstxt.org was last modified on 10 August 2026, and the author reports that thousands of sites publish a file and that Chrome's Lighthouse audits for one.

What nobody has shown is that search products rank or cite pages because the file exists. The OpenAI and Anthropic documents that explain how their crawlers work describe robots.txt tokens, IP ranges and what blocking does, and neither mentions llms.txt as a signal. Google's AI features guide says you do not need to create new machine-readable files, AI text files or markup to appear in AI Overviews or AI Mode.

Where llms.txt does have a clear use is on-demand retrieval: a coding agent or assistant pointed at your documentation can read one small file and follow the links instead of crawling your whole site. That makes it most useful for software documentation, API references and product facts that an agent needs to get right.

## How to write one, step by step
A valid file takes about ten minutes. Treat it as documentation hygiene, not a ranking tactic.

1. **Start with the required H1.** The only required element is an H1 with the name of the project or site, for example # Example Co. Everything else is optional.
2. **Add a one-paragraph summary in a blockquote.** Write two or three factual sentences: what the product is, who it is for, and any fact an agent would get wrong without being told. Put it in a blockquote right under the H1.
3. **List the pages that matter under H2 headings.** Under headings such as ## Docs and ## Pricing, add markdown links in the form - [Page name](https://example.com/page.md): one line of what it contains. Link to clean markdown versions where you have them.
4. **Publish markdown twins for key pages.** The proposal suggests serving a markdown version of a page at the same URL with .md appended. Point the twin's canonical header at the HTML page so the two do not compete in search.
5. **Check that every link works and the facts are current.** A stale pricing line is worse than no file. Fetch each link, confirm a 200 response, and keep one source of truth for facts such as plan limits and tool counts.
6. **Test it with an agent.** The proposal's own advice: ask an assistant questions about your product, giving it only your llms.txt as a starting point, and see whether it answers correctly.

## What is established and what is not
Claims about llms.txt and what the cited sources say (read 13 November 2026)

| Claim | What the source says | Source |
| --- | --- | --- |
| llms.txt is a proposal for a root markdown file | Yes: proposed by Jeremy Howard, published 3 September 2024, modified 10 August 2026 | llmstxt.org |
| It is needed to appear in Google AI Overviews or AI Mode | No: Google says you do not need new machine-readable files, AI text files or markup | Google Search Central |
| OpenAI crawlers use it as a signal | Not stated in OpenAI's crawler documentation | OpenAI crawler docs |
| Anthropic crawlers use it as a signal | Not stated in Anthropic's crawler documentation | Anthropic Help Center |
| Agents fetching documentation benefit from it | That is the use case the proposal describes | llmstxt.org |

## When to use SEOBolt for this
SEOBolt's free llms.txt Checker (/tools/llms-txt-checker) fetches /llms.txt and checks the llmstxt.org format, markdown links and a sample of link status codes, and the free llms.txt Generator (/tools/llms-txt-generator) writes a starting file. The free audit reports a missing llms.txt as a low-priority recommendation next to schema and crawler-access checks, and says plainly that major AI search providers do not document it as a requirement.

SEOBolt itself publishes /llms.txt and markdown twins for its key pages, generated from the same constants as the site so facts such as the tool count stay in step.

MCP tools involved: `seo_issues`.

**When SEOBolt is not the right tool:** If your goal is more citations, spend the time on crawler access, clear answer-first pages and accurate third-party listings first. An llms.txt file will not compensate for blocked bots or thin pages.

## Prompt to give your AI assistant

With the SEOBolt MCP server connected (https://seobolt.app/mcp-seo):

```text
Assign to SEOBolt: use seo_issues to tell me whether my site has an llms.txt and whether its links resolve. If it is missing, draft one from my top pages in the llmstxt.org format. Show me what you did and the draft so I can approve it.
```

## FAQ
### Does ChatGPT read llms.txt?
OpenAI's crawler documentation does not mention llms.txt as a signal. OpenAI publishes an llms.txt for its own developer documentation, which shows the format is useful for documentation, not that ChatGPT search uses yours.

### Can llms.txt hurt my site?
Not by itself. The risk is stale or wrong facts in a file that agents trust, so keep it generated from the same source as your site.

### Where does the file go?
At the site root as /llms.txt, or at a subpath such as /docs/llms.txt to cover only the pages under that path.

### Is llms.txt the same as robots.txt?
No. robots.txt tells crawlers what they may access. llms.txt gives an agent a curated overview when it needs information; the proposal says it can complement robots.txt, not replace it.

## Sources
- [llmstxt.org: The /llms.txt file, v2](https://llmstxt.org/), read 13 November 2026
- [Google Search Central: AI features and your website](https://developers.google.com/search/docs/appearance/ai-features), read 13 November 2026
- [OpenAI: Overview of OpenAI crawlers](https://platform.openai.com/docs/bots), read 13 November 2026
- [Anthropic Help Center: Does Anthropic crawl data from the web, and how can site owners block the crawler?](https://support.claude.com/en/articles/8896518), read 13 November 2026

## Related
- [How do I check whether my site is visible in AI search?](https://seobolt.app/solutions/check-ai-search-visibility-geo)
- [llms.txt Checker (free tool)](https://seobolt.app/tools/llms-txt-checker)
- [llms.txt Generator (free tool)](https://seobolt.app/tools/llms-txt-generator)
- [LLM SEO audit](https://seobolt.app/llm-seo-audit)
- [All problems and solutions](https://seobolt.app/solutions)

HTML version: https://seobolt.app/solutions/do-i-need-llms-txt