> Content index: https://quillly.com/docs/llms.txt
> Canonical page: https://quillly.com/docs/content-types

---
title: Quillly Content Types & Endpoints: Blogs, Docs, More
description: How Quillly's blog, docs, changelog, and tracked-page endpoints work — path prefixes, reverse-proxy serving on your domain, sitemaps, and llms.txt.
keywords: quillly content types, content endpoints, docs on your domain, changelog hosting, llms.txt sitemap
published: 2026-07-11
updated: 2026-07-26
url: https://quillly.com/docs/content-types
word_count: 497
---

# Content types & endpoints

> How Quillly's blog, docs, changelog, and tracked-page endpoints work — path prefixes, reverse-proxy serving on your domain, sitemaps, and llms.txt.

Canonical: https://quillly.com/docs/content-types
Published: 2026-07-11

A Quillly website is a set of **content endpoints** — each one a path on your domain that holds one of four **content types**. This page explains those content types, how their path prefixes work, how pages are served on your domain, and where the sitemaps and llms.txt files live.

## The four content types

- **Blog** — your posts, at `/blogs` by default. Every site has exactly one default blog endpoint.

- **Docs** — documentation pages like this one, at `/docs`.

- **Changelog** — product updates, at `/changelog`. Entries share one timeline page.

- **Pages** — your own existing site pages, such as pricing, features, or landing pages. Quillly *tracks* these for SEO and freshness; it does not author them.

You author blogs, docs, and changelogs with the same content tools — `create_content`, `update_content`, `publish_content` — by passing `content_type: "blog" | "docs" | "changelog"`. There are deliberately no separate docs or changelog tools. "Page" items cannot be authored; they are discovered from your sitemap.

## Path prefixes

Each endpoint maps to a single path segment: `/blogs`, `/docs`, `/changelog`. A "pages" endpoint can sit at the root (`/`). The default blog endpoint mirrors your site's blog path; additional endpoints are independent and additive — adding docs never changes where your blog serves.

## How serving works

For a custom domain, you put a **reverse proxy** in front of each endpoint: a rule that forwards `yourdomain.com/docs/*` (and `/blogs/*`, `/changelog/*`) to your Quillly serve URL. The page renders on Quillly and streams back through your domain, so visitors and search engines only ever see `yourdomain.com`. Quillly verifies each path independently by fetching it and detecting the `<meta name="quillly-site">` tag in the HTML.

![A request to yourdomain.com/docs flows through your reverse proxy to Quillly's serve route and back](https://quillly.com/serve/v1/019c64a2-a62f-7793-aa68-2c78316d3309/images/f080a970041b63182a312a3ec8743ebb7fbf666b.webp)

If you use a free `*.quillly.com` subdomain instead of a custom domain, every endpoint is served from the subdomain root — no proxy needed.

## Sitemaps

Every endpoint serves its own sitemap on its own path:

```
https://yourdomain.com/blogs/sitemap.xml
https://yourdomain.com/docs/sitemap.xml
https://yourdomain.com/changelog/sitemap.xml
```

Blog and docs sitemaps list per-item URLs; the changelog sitemap points at the timeline page. Quillly only advertises a non-default endpoint's sitemap to search engines once that endpoint's proxy is verified and sitemaps are enabled — so engines never get a URL that 404s.

## llms.txt

Quillly generates an `llms.txt` index — the [llmstxt.org](https://llmstxt.org) convention — for AI assistants and crawlers, straight from your live content so it never drifts. Each endpoint exposes one on its prefix:

```
https://yourdomain.com/docs/llms.txt
```

You can also append `/llms.txt` to any individual post URL to get its full text as clean markdown. On a hosted subdomain, a single root `/llms.txt` covers every type.

## Keep it fresh

Tracked pages are re-checked on a schedule, but you can trigger an immediate re-check the moment you deploy — see [Deploy hook: keep tracked pages fresh](/docs/deploy-hook). To connect an AI client and start publishing, see [Connect Claude, ChatGPT or Cursor](/docs/connect-your-ai).