What is an llms.txt file?

llms.txt is a plain Markdown file placed at the root of a website, for example yourdomain.com/llms.txt. It gives AI assistants and coding agents a short, curated map of the site: a title, a summary, and links to the most useful pages with a one-line note on each. It was proposed in 2024 as a community convention and is not an official web standard.

Where do I put the llms.txt file?

Upload it to the root of your domain so it loads at https://yourdomain.com/llms.txt. Save it as plain text with UTF-8 encoding and the exact lowercase file name llms.txt. If your site uses a subdomain, put a separate file at the root of that subdomain.

Does Google use llms.txt?

No. Google has said that Search does not use llms.txt, including for AI Overviews and AI Mode, and its AI optimization guidance says you do not need special AI files to appear there. Adding one will neither help nor hurt your Google rankings.

Will llms.txt improve my AI citations or rankings?

There is no proven benefit. Independent studies in 2025 and 2026 found no clear link between having an llms.txt file and being cited by AI tools, and no major AI provider has confirmed that its crawler relies on the file. Some AI tools and coding agents do read it, so treat it as low-cost documentation for those tools rather than a ranking tactic.

What is the difference between llms.txt and robots.txt?

robots.txt controls which paths crawlers are allowed to access, and well-known crawlers respect it. llms.txt is a curated reading list that suggests which pages are most useful. It does not block or allow anything, so it cannot replace robots.txt or a sitemap.

What is llms-full.txt?

llms-full.txt is an optional companion file that puts the full text of your key pages into one Markdown file, so a tool can read everything in a single request. llms.txt only lists links with short notes. Large sites often skip the full version because of its size.

What should I include in my llms.txt?

List your most useful pages: documentation, guides, tools, pricing and policies. Give each link a short, specific note. Keep the file short, put the most important links first, and move nice-to-have pages into a section named Optional so tools with limited context can skip them.

Can I add llms.txt on Blogger, Shopify or other hosted platforms?

Some hosted platforms do not let you upload a file to the site root. If that is your case, serve /llms.txt through a redirect or proxy rule, for example with a CDN such as Cloudflare, that points to a plain-text file hosted elsewhere. Open yourdomain.com/llms.txt in a browser to confirm it returns plain text.

Is this llms.txt generator free and private?

Yes. It is free, needs no signup, and runs entirely in your browser. Your site details and links are never uploaded. The Share link button puts your form data in the URL, so only share that link with people you trust.

A Markdown file at /llms.txt that gives language models a curated map of a site: a title, a short summary, and annotated links to the most useful pages.

What is the difference between llms.txt and sitemap.xml?

sitemap.xml lists every URL for search crawlers. llms.txt is a short, curated, annotated list that a model can read in one pass. It complements the sitemap and does not replace it.

What is the llms-full.txt file?

/llms.txt is an index of links. /llms-full.txt is a common companion convention, not part of the core proposal, that holds the full text of key pages in one file. The aggregated Markdown content of your key pages in one file, with a heading and source URL for each page, and no navigation, footer or other boilerplate.

What is the exact syntax for links in llms.txt?

- [Title](https://example.com/page): Description The link is required and the note after the colon is optional.

Where should the llms.txt file be hosted?

/llms.txt at the root of the host, for example https://example.com/llms.txt. The proposal also allows a file in a subpath for a docs section.

Can AI tools read llms.txt if bots are blocked in robots.txt?

Compliant crawlers will not fetch /llms.txt if robots.txt disallows it for their user agent. A tool that a person points directly at the file or its text can still read it.

What is the purpose of the blockquote summary in llms.txt?

It gives a short summary with the key facts a model needs to read the rest of the file, so keep it factual and specific.

How do AI agents like Cursor or Claude use llms.txt?

Cursor can index documentation from a URL. I could not confirm that it reads /llms.txt automatically, so check Cursor’s documentation and test it with your own link. Claude does not fetch /llms.txt on its own. Upload the file or paste its contents as project knowledge, or give Claude the URL when web access is on.

LLMs.txt Generator: Create AI Context Files

llms.txt Generator: Create AI Context Files

llms.txt generator

Save on Google

Free online llms.txt generator. Add your site name, summary and links, then copy or download a ready-to-upload file. A sample is preloaded: edit it, or press Reset to start blank. Everything runs in your browser. Nothing is uploaded.

Configure llms.txt

Site or project name Summary (one or two sentences) Extra details (optional)

Results & Output

What is llms.txt?

llms.txt is a plain Markdown file at yourdomain.com/llms.txt that gives AI assistants a short, clean map of your site: a title, a summary, and links to your best pages with a note on each. It is a community proposal, not an official web standard, and support is voluntary. Google says its Search and AI features do not use it, so treat it as documentation for the AI tools and coding agents that do read it.

llms.txt technical reference: 60 questions developers search

Rules of thumb are labelled as such. The proposal itself is documented at llmstxt.org by Answer.AI. Items I could not confirm are marked that way.

60 of 60 shown

Technical Specification / Developer Query	Direct Technical Specification (Answer & Code)	Category / Scope
What is llms.txt?	A Markdown file at `/llms.txt` that gives language models a curated map of a site: a title, a short summary, and annotated links to the most useful pages.	Core / Origin
Full form of llms.txt	“llms” stands for large language models (LLMs). The file is a reading guide written for them.	Core / Origin
Primary purpose vs traditional sitemap.xml	`sitemap.xml` lists every URL for search crawlers. `llms.txt` is a short, curated, annotated list that a model can read in one pass. It complements the sitemap and does not replace it.	Core / Origin
Who proposed the standard	Jeremy Howard of Answer.AI proposed it in September 2024. The proposal is documented at llmstxt.org.	Core / Origin
Official standard status	A community proposal. It is not published by the W3C or IETF, there is no formal standards process, and support by AI tools is voluntary.	Core / Origin
Standard root file path	`/llms.txt` at the root of the host, for example `https://example.com/llms.txt`. The proposal also allows a file in a subpath for a docs section.	Core / Origin
Supported MIME type	The proposal does not mandate one. Serve it as `text/plain; charset=utf-8` or `text/markdown; charset=utf-8` so browsers show the text instead of downloading it.	Core / Origin
Default character encoding	UTF-8. Declare it with `charset=utf-8` in the `Content-Type` header.	Core / Origin
Case sensitivity of filename	Use lowercase `llms.txt`. Many servers treat file names as case sensitive, so `LLMS.TXT` may return 404.	Core / Origin
HTTP status code for missing file	`404 Not Found`. Make sure a missing file does not return an HTML page with `200`, because a tool may read that page as the file.	Core / Origin
H1 title requirement	One `# Title` line at the top is the only required element. Load sample template	Markdown Syntax
Blockquote summary syntax	`> One or two sentence summary` placed right after the title. Load sample template	Markdown Syntax
Purpose of blockquote summary	It gives a short summary with the key facts a model needs to read the rest of the file, so keep it factual and specific.	Markdown Syntax
Narrative paragraphs placement	Optional paragraphs or other non-heading Markdown go after the summary and before the first `##` section.	Markdown Syntax
H2 section headers syntax	`## Section Title` starts a section that holds a list of links.	Markdown Syntax
Recommended standard section names	Common choices are `## Docs`, `## Documentation`, `## API Reference`, `## Guides` and `## Examples`. Only `## Optional` has a defined meaning.	Markdown Syntax
Link list item syntax	`- [Title](https://example.com/page): Description` The link is required and the note after the colon is optional.	Markdown Syntax
Requirement for link descriptions	Optional but recommended. A specific note tells a model what the page covers before it loads it.	Markdown Syntax
Absolute vs relative URLs policy	Use absolute `https://` URLs so a tool can fetch each link without guessing the base domain.	Markdown Syntax
Multiple H1 tags handling	Use exactly one `# Title`. A second H1 can confuse parsers, so use `##` for sections. The generator above writes a single H1. Open the generator	Markdown Syntax
Difference between llms.txt and llms-full.txt	`/llms.txt` is an index of links. `/llms-full.txt` is a common companion convention, not part of the core proposal, that holds the full text of key pages in one file.	Advanced Directives
What should llms-full.txt contain?	The aggregated Markdown content of your key pages in one file, with a heading and source URL for each page, and no navigation, footer or other boilerplate.	Advanced Directives
Handling secondary links	Put links a tool can skip under `## Optional`. The proposal says they can be left out when a shorter context is needed.	Advanced Directives
Linking raw Markdown docs vs rendered HTML	Link to clean Markdown versions when you have them, for example `page.md`. They carry less markup and use fewer tokens than HTML.	Advanced Directives
Stripping navigation, footer and boilerplate links	List only pages with real content. Leave out menus, cookie notices, login, cart and footer links.	Advanced Directives
Maximum recommended file size for /llms.txt	There is no official limit. As a rule of thumb, keep `/llms.txt` under about 50 KB.	Advanced Directives
Maximum recommended token budget	There is no official limit either. A practical target is under about 10,000 tokens, and roughly 4 characters equal 1 token.	Advanced Directives
Preventing LLM hallucinations via precise titles	Use exact page, product and function names in titles and notes so a model has less to guess. This lowers the risk but gives no guarantee.	Advanced Directives
Handling non-standard HTTP redirects	Serve the file directly with `200`. If you must redirect, use one `301` hop to the final text file, and avoid `302` chains or redirects to HTML pages.	Advanced Directives
Handling authenticated or private documentation	Do not list it. The file is public, and crawlers cannot open links that need a login.	Advanced Directives
Nginx header configuration for text/markdown	`location = /llms.txt { default_type text/markdown; charset utf-8; }`	Server and Network
Apache .htaccess directive for /llms.txt	`<Files "llms.txt"> ForceType "text/markdown; charset=utf-8" </Files>`	Server and Network
Cache-Control header recommendation	`Cache-Control: public, max-age=86400` caches the file for one day. Use a shorter time if you update it often.	Server and Network
Cloudflare edge caching rules for llms.txt	Create a Cache Rule for URI path equal to `/llms.txt`, set it eligible for cache with a one-day Edge TTL, and purge it after each edit.	Server and Network
Cross-Origin Resource Sharing (CORS)	`Access-Control-Allow-Origin: *` is only needed if browser-based tools on other origins fetch the file. Crawlers and server-side tools do not need it.	Server and Network
Robots.txt relationship	`robots.txt` controls crawler access and `llms.txt` controls nothing. Keep `/llms.txt` allowed in `robots.txt` so compliant crawlers can fetch it.	Server and Network
Can crawlers read llms.txt if root is disallowed in robots.txt?	Compliant crawlers will not fetch `/llms.txt` if `robots.txt` disallows it for their user agent. A tool that a person points directly at the file or its text can still read it.	Server and Network
Gzip and Brotli compression support	Yes. Servers can compress it with `Content-Encoding: gzip` or `br`, and clients decompress it transparently.	Server and Network
Staging vs Production llms.txt handling	Keep staging private behind authentication. Make sure production links use the production domain, not staging URLs.	Server and Network
Automated CI/CD deployment via GitHub Actions	Add a build step such as `run: node scripts/build-llms.js` that writes `public/llms.txt` before the deploy step.	Server and Network
Cursor IDE integration with llms.txt	Cursor can index documentation from a URL. I could not confirm that it reads `/llms.txt` automatically, so check Cursor’s documentation and test it with your own link.	AI Agents and IDEs
Windsurf / Codeium integration mechanism	I could not confirm a built-in `llms.txt` mechanism. You can add your docs URL or paste the file content as context, and check Windsurf’s documentation for current support.	AI Agents and IDEs
Claude Projects custom context integration	Claude does not fetch `/llms.txt` on its own. Upload the file or paste its contents as project knowledge, or give Claude the URL when web access is on.	AI Agents and IDEs
OpenAI Custom GPTs knowledge base ingestion	As far as I know, Custom GPTs do not discover `/llms.txt` automatically. Upload `llms-full.txt` or your key docs as knowledge files.	AI Agents and IDEs
LangChain document loader for llms.txt	I am not aware of a dedicated loader. Load the URL with a generic loader such as `WebBaseLoader`, then split on `##` headings with `MarkdownHeaderTextSplitter`.	AI Agents and IDEs
LlamaIndex reader compatibility	There may be no dedicated reader. Load the URL with `SimpleWebPageReader`, or download the file and read it with `SimpleDirectoryReader`.	AI Agents and IDEs
Perplexity and live search engines parsing	No major search or answer engine has confirmed that it relies on `llms.txt`, and Google says its Search does not use it. Treat it as optional.	AI Agents and IDEs
Semantic chunking behavior on llms.txt files	It depends on the tool. A small file may be read whole or split at `##` headings, so write each section to make sense on its own.	AI Agents and IDEs
Token consumption efficiency vs sitemap scraping	A short `llms.txt` costs a few thousand tokens. Crawling many HTML pages from a sitemap costs far more and adds no context. The ratio varies by site.	AI Agents and IDEs
Formatting code documentation for agent tool-calling	Link to Markdown reference pages and put the function or endpoint name in the title: `- [create_invoice](https://example.com/api/create_invoice.md): POST /invoices, parameters and errors`	AI Agents and IDEs
Validating syntax with online validators	Run the file through an llms.txt validator, then open every link. A validator checks structure, not whether links load. Open the generator	Validation and Maintenance
Impact of trailing slashes in URLs	`/docs` and `/docs/` can be different URLs. List the exact canonical form your site serves with `200`, so models do not hit a redirect.	Validation and Maintenance
Handling non-ASCII or special characters in link titles	UTF-8 titles are fine. Percent-encode spaces and brackets in URLs, for example `%20`, `%28` and `%29`.	Validation and Maintenance
Broken link auditing and dead link removal frequency	Check links on every release and at least monthly as a rule of thumb. Fix or remove any `404` or redirected link.	Validation and Maintenance
Auto-generating llms.txt from Next.js / Astro static routes	Write the file at build time into `public/llms.txt`. In Astro you can also export a `GET` function from `src/pages/llms.txt.ts`. Open the generator	Validation and Maintenance
Auto-generating llms.txt from existing sitemap.xml	Read each `<loc>`, group URLs by first path segment into `##` sections, take titles from page titles, then edit by hand, because a sitemap has no descriptions.	Validation and Maintenance
Formatting API endpoints (OpenAPI / Swagger mapping)	Link the spec and per-tag reference pages: `- [OpenAPI spec](https://api.example.com/openapi.json): Full REST schema`. Group endpoints by tag.	Validation and Maintenance
Handling versioned documentation (/v1/ vs /v2/)	Point main sections to the current stable version. Put older versions such as `/v1/` under `## Optional` and name the version in each title.	Validation and Maintenance
Multi-language site handling	No standard exists. Either use one file with a section per language or serve files by path, such as `/es/llms.txt`. Avoid inventing names like `llms-es.txt`.	Validation and Maintenance
Keeping llms.txt synchronized with code releases	Generate it in CI on each release, validate it, and fail the build if a listed link returns `404`.	Validation and Maintenance

llms.txt FAQ

Question	Answer
What is an llms.txt file?	llms.txt is a plain Markdown file placed at the root of a website, for example yourdomain.com/llms.txt. It gives AI assistants and coding agents a short, curated map of the site: a title, a summary, and links to the most useful pages with a one-line note on each. It was proposed in 2024 as a community convention and is not an official web standard.
Where do I put the llms.txt file?	Upload it to the root of your domain so it loads at https://yourdomain.com/llms.txt. Save it as plain text with UTF-8 encoding and the exact lowercase file name llms.txt. If your site uses a subdomain, put a separate file at the root of that subdomain.
Does Google use llms.txt?	No. Google has said that Search does not use llms.txt, including for AI Overviews and AI Mode, and its AI optimization guidance says you do not need special AI files to appear there. Adding one will neither help nor hurt your Google rankings.
Will llms.txt improve my AI citations or rankings?	There is no proven benefit. Independent studies in 2025 and 2026 found no clear link between having an llms.txt file and being cited by AI tools, and no major AI provider has confirmed that its crawler relies on the file. Some AI tools and coding agents do read it, so treat it as low-cost documentation for those tools rather than a ranking tactic.
What is the difference between llms.txt and robots.txt?	robots.txt controls which paths crawlers are allowed to access, and well-known crawlers respect it. llms.txt is a curated reading list that suggests which pages are most useful. It does not block or allow anything, so it cannot replace robots.txt or a sitemap.
What is llms-full.txt?	llms-full.txt is an optional companion file that puts the full text of your key pages into one Markdown file, so a tool can read everything in a single request. llms.txt only lists links with short notes. Large sites often skip the full version because of its size.
What should I include in my llms.txt?	List your most useful pages: documentation, guides, tools, pricing and policies. Give each link a short, specific note. Keep the file short, put the most important links first, and move nice-to-have pages into a section named Optional so tools with limited context can skip them.
Can I add llms.txt on Blogger, Shopify or other hosted platforms?	Some hosted platforms do not let you upload a file to the site root. If that is your case, serve /llms.txt through a redirect or proxy rule, for example with a CDN such as Cloudflare, that points to a plain-text file hosted elsewhere. Open yourdomain.com/llms.txt in a browser to confirm it returns plain text.
Is this llms.txt generator free and private?	Yes. It is free, needs no signup, and runs entirely in your browser. Your site details and links are never uploaded. The Share link button puts your form data in the URL, so only share that link with people you trust.