# Squarespace llms.txt: How to Add One That Works (2026)
Author: Sai Krishna
Author URL: https://superblog.ai/blog/author/sai-krishna/
Published: 2026-09-15
Meta Title: Squarespace llms.txt: How to Add One (2026)
Meta Description: Squarespace 7.1 has a native llms.txt field, off by default. The exact setup path, a paste-ready template, the 7.0 workaround, and what the field misses.
Tags: SEO, no-code, ai search
Tag URLs: SEO (https://superblog.ai/blog/tag/seo/), no-code (https://superblog.ai/blog/tag/no-code/), ai search (https://superblog.ai/blog/tag/ai-search/)
URL: https://superblog.ai/blog/squarespace-llms-txt/

![Squarespace llms.txt](https://prod.superblogcdn.com/site_cuid_ckox4in4f002nl8lhcib41g2u/images/squarespace-llms-txt-1789124443522-compressed.png)

Squarespace llms.txt is a native setting on version 7.1, and it is switched off until you fill it in yourself. Open the SEO/AI Visibility panel, go to SEO Settings, open the LLMS.txt tab, type the contents of the file into the text field, and save. That is the entire feature. Squarespace publishes the file for you. It does not write the file, and it does not touch the file again when you publish a post.

Below: the exact path, a paste-ready template, the separate AI-crawler checkbox that can quietly undo the work, the 7.0 fallback, and what to do when your blog is the part machines should be reading.

## Where the llms.txt setting lives on Squarespace 7.1

Verified against Squarespace's own [Create an llms.txt file](https://support.squarespace.com/hc/en-us/articles/47434125611277-Create-an-llms-txt-file) documentation on September 11, 2026:

1. Open the **SEO/AI Visibility** panel.
2. Click **SEO Settings**.
3. Open the **LLMS.txt** tab.
4. Paste your file contents into the text field.
5. Save.

To remove it later, return to the same tab, choose **Delete**, then **Save**.

Two details matter before you start. The feature is version 7.1 only. And Squarespace is candid that the contents are your problem: "We can't offer specific advice on creating text for your llms.txt file, as it's a proposed standard that's still actively in development." The help page refers you to llmstxt.org, or to hiring a Squarespace Expert. You need neither. The format is short and the structure is fixed.

## A paste-ready llms.txt template for a business site

An llms.txt file is markdown with four moving parts: a site name as an H1, a one-line description as a blockquote, H2 sections grouping your links, and bullet links with a short description after each one. Replace the example values below and you have a working file.

```
# Acme Analytics

> Acme Analytics is revenue analytics for B2B SaaS teams. This file indexes the pages that explain what we do, how the product works, and what we publish.

## Core pages

- [Product overview](https://acme.com/): What Acme does and who it is for.
- [Pricing](https://acme.com/pricing): Plans, limits and what each tier includes.
- [About](https://acme.com/about): Company background, founders and location.

## Blog

- [Blog index](https://acme.com/blog): Every article, newest first.
- [How to measure net revenue retention](https://acme.com/blog/net-revenue-retention): The formula, benchmarks by segment, and the three ways teams get it wrong.
- [Attribution models compared](https://acme.com/blog/attribution-models): First touch, last touch and multi touch, with the case for each.

## Contact

- [Contact](https://acme.com/contact): Sales and support.

```

Keep it to the pages you would send a new customer. Twenty well-chosen links beat two hundred, because the value of the file is curation, not coverage. Write each description as a fact about the page rather than a sales line.

To skip the assembly, our free [llms.txt generator](/tools/llms-txt-generator) reads your sitemap.xml, pulls titles and descriptions, and drafts the file. Squarespace publishes a sitemap automatically, so it works on any 7.1 site. Paste the draft into the field, cut it to the pages that matter, and save.

## What the Squarespace llms.txt field does not do

The setting is a text box, and it behaves like one.

**It never updates itself.** Publish a post and the field still says what it said last month. Every new article has to be typed in by hand. On a weekly cadence the file is stale within a month and actively misleading within three, because it now presents your older posts as your important ones.

**There is no markdown copy of your pages.** The llmstxt.org proposal expects the links in the file to resolve to clean markdown wherever possible, so a model reads prose instead of parsing a page full of layout markup. Squarespace serves the rendered HTML page and nothing else, so the index points at documents that still have to be stripped down before they can be read.

**It is one file for the whole site.** There is no blog-aware behaviour, no per-section file, and no check that the URLs you typed still exist. Change a post slug and the file quietly points at a 404.

None of this makes the feature worthless. A hand-written llms.txt for a ten-page brochure site is a genuinely good use of ten minutes, because that site changes twice a year. The mismatch appears when the thing you publish most is the thing the file cannot keep up with.

## The AI crawler checkbox is a separate switch, and a blunt one

Under **Settings > Crawlers**, Squarespace offers a checkbox labelled "Block known artificial intelligence crawlers". It is unchecked by default, and it works by writing entries into your robots.txt. Per Squarespace's [AI crawler exclusion documentation](https://support.squarespace.com/hc/en-us/articles/360022347072-Request-that-AI-models-exclude-your-site), read September 11, 2026, ticking it blocks 26 named crawlers: AI2Bot, Ai2Bot-Dolma, aiHitBot, Amazonbot, anthropic-ai, Applebot-Extended, Bytespider, CCBot, ClaudeBot, cohere-ai, cohere-training-data-crawler, DuckAssistBot, FacebookBot, Google-Extended, GoogleOther, GoogleOther-Image, GoogleOther-Video, GPTBot, img2dataset, Meta-ExternalAgent, MyCentralAIScraperBot, omgili, omgilibot, Quora-Bot, TikTokSpider and YouBot.

Read the list before you tick the box, because it is all or nothing. You cannot keep Google-Extended while refusing Bytespider, or allow ClaudeBot while blocking CCBot. One checkbox covers all 26.

Two things are worth knowing. OAI-SearchBot and ChatGPT-User are not on the list, so the checkbox does not stop ChatGPT fetching a page while answering a question about it. And Squarespace states the obvious limit plainly: "updating robots.txt this way is only a request, and malicious crawlers may still misuse your content". They also note it does not retroactively pull anything already scraped.

Deciding which of these crawlers you actually want is a separate exercise from writing the file, and our [guide to generative engine optimization](/geo) walks through the crawler taxonomy in full.

## Adding llms.txt to a Squarespace 7.0 site

Version 7.0 has no LLMS.txt tab, so the older community workaround is still the only route. It places the file by redirection rather than by writing to the root, which you cannot reach on a hosted platform:

1. Write the file and upload it through **Link Manager**, which makes it reachable at `yoursite.com/s/llms.txt`.
2. Trace that URL to find the static asset address Squarespace's CDN actually serves it from. The `/s/` path answers with a 302, so you want the destination, not the hop.
3. In **Settings > Developer Tools > URL Mappings**, add a 301 from `/llms.txt` to that CDN address.

The [beyondspace.studio walkthrough](https://www.beyondspace.studio/blog/how-to-add-llmstxt-to-squarespace) documents the method in detail, and now carries its own note that Squarespace shipped native support on July 27, 2026. Treat the hack as the 7.0 answer only. If you are on 7.1, use the field.

## A blog where llms.txt writes itself

If the pages you most want machines to read are blog posts, the honest conclusion is that the file should be generated by whatever publishes them. That is what [Superblog](https://superblog.ai) does, and you do not have to leave Squarespace to get it. Superblog can take over the `/blog` path on the domain you already own, using one routing rule at Cloudflare that sits in front of it. Everything you built in Squarespace stays where it is, and nobody ever leaves your address bar. Prefer to keep the blog on its own hostname? A subdomain or a second custom domain is included at every tier, SSL certificate and all. Walkthrough: [blog for Squarespace](/use-cases/blog-for-squarespace).

Where your llms.txt comes fromWho writes itWhat happens when you publish a postSquarespace 7.1 fieldYou, by hand, into a text boxNothing. The file says what it said last time you edited itSquarespace 7.0 redirect hackYou, by hand, plus a Link Manager upload and a 301Nothing, and you re-upload the file to change itSuperblogGenerated for youThe file regenerates on that deploy, and the new post is also served as markdown

On the specific problem this post opened with: llms.txt is a toggle in Settings > SEO, and once it is on, the file is rebuilt on every deploy with every post in it. Alongside it, each post is also served as plain markdown at its own URL plus `.md`, and the llms.txt index points at those markdown copies. That is the full readability stack: a server-rendered page at a real URL, a clean markdown mirror of the same post, and an index that stays current without anyone opening a settings tab. It is how you give AI engines something to find and cite rather than hoping they parse your layout.

The rest of the automation runs on the same schedule. Hitting publish writes the technical markup for you: a canonical tag, Open Graph metadata, and JSON-LD blocks for Article, WebPage, Organization, ImageObject and FAQ. Your XML sitemap is rebuilt during the deploy, and IndexNow pings the moment the post is live. Publishing in several languages brings hreflang plus one sitemap per language, from Super upward.

Around the SEO layer:

- **Nothing to maintain.** No servers, no plugins, no update queue, no security patches. Posts compile down to static files served from a global CDN, so opening one queries nothing and executes nothing. That is the entire security story.
- **Writing and review.** Autosave, co-authors on a single post, scheduled publishing and collaborative review from Pro, and Admin, Editor and Writer roles with 5 team members on Pro, 10 on Super, and 25 on Business.
- **Analytics with no cookie banner.** Pirsch reports visits without setting a cookie, from Pro upward. Every plan also supports Google Analytics if you already run it.
- **Business plan extras.** The dashboard runs keyword research against live DataForSEO data, returning volume, competition and CPC for 25 keywords at a time. It will also build a featured image and an Open Graph image out of the post itself, one credit apiece.
- **Speed you never tune.** A guaranteed 90+ Lighthouse score on every page, WebP conversion handled automatically, and an edge network of 200+ locations.

Moving the archive needs honest expectations. Squarespace is missing from the list of direct importers, and the reason sits in Squarespace's own export format: what it hands you is WordPress-flavoured XML, not anything Squarespace-specific. Imports by URL or API token cover WordPress, Ghost, Webflow, Wix, Shopify, Blogger, Drupal and Docusaurus, while Markdown, HTML, docx, ZIP and CSV arrive as file uploads. So a departure starts with that XML file and a redirect map. Settle the URL mapping before you migrate anything.

Scale is rarely the blocker. One finance publication brought over 15,000 posts off WordPress and onto Superblog. Another 23 blogs on the platform sit at 1,000 posts or more, and upward of 150,000 posts are served altogether. Two names you can look up: PrintStop shifted its business blog away from WordPress, and Elephas skipped WordPress entirely and started here.

Pro runs $49 a month, holds 1,000 posts and seats 5 people. At $99, Super lifts the post cap under fair use, and adds multilingual SEO plus API access. Trying it takes 7 days and no card. There is a 30-day money-back guarantee. Plan-by-plan numbers live on the [pricing page](/pricing).

## Squarespace llms.txt questions people also ask

**Does Squarespace generate the llms.txt file for me?**

No. The 7.1 feature gives you a text field and publishes whatever you type into it. There is no generation step and no automatic refresh when you add a page or a post.

**Do I still need llms.txt if I have a sitemap?**

They do different jobs. A sitemap lists every URL for indexing. An llms.txt is a curated index that says what the site is and which pages matter, written in markdown a model can read directly. Keep both.

**Does blocking AI crawlers cancel out my llms.txt?**

Largely, yes, for the crawlers on the list. If robots.txt refuses GPTBot and ClaudeBot, a file written for them to read is not going to be fetched by them. Decide which crawlers you want before you write the file.

**Can I keep my Squarespace site and move only the blog?**

Yes. The blog runs at `yoursite.com/blog` through a reverse proxy rule while every Squarespace page stays untouched. The wider case for and against the platform's blog is in [is Squarespace good for blogging](/blog/is-squarespace-good-for-blogging/).

## Start with the file, then look at the blog

If you are on 7.1, fill the field today. The template above turns ten minutes into a real llms.txt, and a brochure site will not need to revisit it for months. If you publish weekly, watch how long the file stays accurate. That number is the argument for moving the blog somewhere that regenerates it.

[Start a free 7-day trial](https://write.superblog.ai/register) and see what the blog looks like when llms.txt, sitemaps, schema and IndexNow all update themselves.


---
This blog is powered by Superblog. Visit https://superblog.ai to know more.
---