Skip to content

Content engine

Link only to pages that exist

Generated articles pick their internal links from your site's last audit crawl. You set the count from 1 to 6, and that count is enforced after generation.

Told to add internal links, an AI writer will invent relative paths, and every invented path ships a 404 in your name.

The other failure is stuffing: six links wedged into a short post because a template asked for six.

  • 1-6internal links per article, and the count is yours to set
  • 3links by default, which suits most article lengths
  • 200pages in the crawl that builds the pool of real targets

Targets come from the crawl, not the model

With a website selected and a completed audit, the writer picks targets from URLs the crawl found on your site. A fresh audit refreshes the pool after a restructure.

Help centre: Internal links
The link view for searchblueprint.io: 44 crawled pages today against a 13 page plan carrying 17 navigation links, 21 recommended links and 0 orphan pages, under the line saying the recommended links are laid over the proposed structure.
Every box in the diagram below is a page the crawl reached, and the recommended links are drawn between those boxes rather than invented as paths.
  • A slider sets the count, 1 to 6

    Three is the default and suits most articles. Go higher on pillar pages that genuinely reference many of your URLs, lower on short posts where six would read as stuffing.

  • The limit is enforced outside the model

    After generation the article is trimmed to your selected count as a hard rule, so the number you set is the number you get.

  • Fewer links when targets do not fit

    The writer links only where a target fits the sentence, so a niche article on a small site may carry fewer than the slider allows. The summary reports the count used.

  • Paste your own targets before the first crawl

    Without a website selected, a manual textarea takes the URLs you want considered, one per line, and the writer picks from those.

  • The audit names your orphan pages

    The audit's links category finds orphan pages nothing points to, broken links, and how link equity flows, so you know which pages need links before you generate.

How it works

The sequence, start to finish. Read it in order: each step assumes the one above it has already run.

  1. Audit the site once

    The crawl builds the pool of real URLs the writer can link to, and its links category shows where you stand.

  2. Set the slider

    Pick your website in Generate, set internal links anywhere from 1 to 6, and generate.

  3. Check the summary

    The summary under the draft counts the internal links used, so you confirm the slider was honored without hunting through the text.

The proposed structure for searchblueprint.io drawn as a page tree, a Homepage box above its section pages, with the recommended internal links laid over it as dashed arrows under a legend saying an arrow points at the page being linked to.
Drawn, not listed. A recommended link is a line between two boxes the plan already contains, so what you read is the structure those links would build rather than a table of pairs.

The other ways people do this

What each approach does, and what happens here instead. Mechanism only, never an outcome claim.

Linking by handHow it worksYou hold the site's URL list in your head and add links while editing, which works until the site outgrows one person's memory.HereTargets come from the URLs your last audit crawl actually found, so the pool is the site as it stands rather than as anyone remembers it.
An AI writer told to add linksHow it worksAsked for six internal links it produces six, inventing relative paths once the real ones run out.HereThe writer picks only from crawled URLs, and the count is trimmed after generation as a hard rule outside the model.
A linking plugin in your CMSHow it worksIt suggests links inside the editor from the posts on that install, for the platform it was built for.HereThe pool comes from a crawl of the live site on any stack, and manual mode takes URLs you paste one per line.
A link-audit crawlerHow it worksIt reports where links are missing and which pages are orphaned, then leaves the writing of those links to you.HereThe audit's links category names the orphans and broken links, and the writer places links from that same crawl.

Try the job, or learn the subject

The free tools run the same checks without an account. The lessons explain the mechanism behind them.

Free tools, no signup

On-page SEO scorer
Score a page's on-page SEO in seconds.
On-page meta checker
Inspect a page's title, meta, headings, canonical and status.
AI content writer
Draft SEO content from a topic or keyword.

Lessons in the Academy

Site architecture and internal links5 min · Beginner
How to organise pages and links so authority flows to the pages that make money.
Internal links to money pages5 min · Intermediate
Link from guides and blog posts to the collections and products they talk about, deliberately.
Topic clusters: own a subject, not a keyword5 min · Intermediate
Group pages around one subject so Google and AI treat you as the authority on it.

Questions people ask

The ones that decide whether this fits, answered with the verdict first.

Can the writer link to a page that doesn't exist?
No, not with a completed audit selected. Targets come from URLs the crawl found, so the one exposure is manual mode with an empty textarea.
How many internal links should I set?
Three, the default, for most articles. Long pillar pieces can carry more, and short posts read better with fewer.
Why did I get fewer links than the slider asked for?
The count is a maximum. The writer links only where a target is relevant, and a niche article on a small site can exhaust its good targets early.
Why do links point at pages I removed?
The pool reflects the last crawl. After a restructure it stays stale until a fresh crawl replaces it, so re-run the audit and generate again.
Does it edit the pages already on my site?
No. Links are placed inside the article the generator produces, before you publish it. Nothing is written into pages that already exist.
What does this add to the cost of an article?
Nothing. Targets come from the audit crawl you already ran, and linking is part of generation, which is one credit for a standard draft.
Do I need a developer to set it up?
No. Run an audit once so the pool exists, pick the website in Generate, then set the slider. With no crawl yet, paste target URLs into the manual textarea.

Where this fits

The jobs either side of this one, and the two pages that read it from your side of the desk.

  • AI article writer

    Give it a keyword. It researches, drafts with internal links and images, then scores the result on the same 36 checks that grade audited pages.

  • Autoblog

    Plan a month of posts. Auto-write drafts each at midnight on its calendar date through the same researched pipeline you run by hand, and every publish is checked at the public URL.

  • Technical SEO audit

    One crawl grades your site A+ to F across eight categories and turns every failure into a fix with its URLs attached.

Content engine

See it on your own site

A free audit grades your site A+ to F, then shows where this feature pays first. No card required.