Search engines see you

What's on from day one#

Nothing to switch on: every page on a docsbook.io address carries all of this, on every plan.

• Server-rendered HTML — the full text is in the first response, readable without running JavaScript
• Title and description — frontmatter title and description, else the # H1 and the opening paragraph; your site name is added once
• One canonical URL per page — your custom domain when you have one
• hreflang — only for the languages this page is actually translated into
• sitemap.xml — every page and every real translation, with lastmod from the last commit that touched the file
• robots.txt — announces your sitemap and lets every crawler in; none is refused by default
• Social cards — Open Graph and X summary_large_image tags with a generated 1200×630 image per page
• JSON-LD — Organization, TechArticle with dates from your git history, and BreadcrumbList on every page
• FAQPage and HowTo — added only when the page has question headings or a "How to…" procedure
• Clean URLs — guides/setup.md is served as …/guides/setup, and the root README.md is the home page
• 308 redirects — a moved page keeps its old address working through .docsbook/redirects.json
• Cached pages — readers and crawlers get a cached copy, replaced when you publish through Docsbook

How do I control what search engines see?#

Set the search title and description, or keep one page out of the index, in the page's frontmatter:

---
title: "Configure a webhook"
description: "Register a webhook, choose its events, and verify the first delivery."
noindex: true
---

noindex: true (or robots: noindex) serves that page with noindex, follow and keeps it readable. To take a whole project out of search, switch off Listed in search results on Settings ▸ Access ▸ Search engines: every page gets noindex, the project leaves the sitemap, and search crawlers are refused in robots.txt.

What the agent does on its own#

On Pro, the agent runs these loops from ready-made triggers. Each one reads a signal, checks it against the rules of the expertise catalog, and changes the docs in a pull request.

Before it writes, the agent researches what people actually search for:

• Search demand — Google Keyword Planner volume, cost-per-click and a 12-month series for up to 25 phrases
• Real phrasing — the questions Google autocompletes after up to 10 seed phrases
• Direction — three months of search interest and the queries rising fastest around a topic
• Results pages — Google's results, People Also Ask and AI Overview, and Bing's results for the same query
• Links — how many domains link to a site, and from which pages
• Competitor docs — up to 50 pages of a rival's documentation, read as clean Markdown
• Questions asked elsewhere — Reddit, Hacker News, Stack Overflow and Discourse threads about your product

A change that claims to move a number states its bet. The agent records the reading it expects to move — Google's position, impressions or clicks for a page, or its views — with today's number, the predicted one and a check date. On that date it takes the same reading again, and Docsbook computes the verdict from the two: as predicted, no effect, went backwards or cannot tell.

See it working#

Analytics ▸ SEO is one list with a View switch. Rows come from Google Search Console and from the checks the agent runs.

Analytics ▸ Audit shows where your pages stand against the three SEO topics of the catalog: Search demand & intent; Crawl, index & speed; Authority & risk.

Tell your agent#

Say what you want in a sentence, from Claude Code, Cursor, Codex or the panel chat:

Find the pages sitting on page two of Google and fix their titles and descriptions.
Why did docs traffic drop last week? Fix the part that is ours.
What do people search for that our docs don't answer? Write the top three pages.
Check every link on the site and fix the broken ones.
Run the organic search audit every week.

FAQ#

{ "version": 1, "redirects": [{ "from": "guides/old-name", "to": "guides/new-name" }] }

Next steps#