Skip to the content
Weekly. On from the first day.

SEO Agent

Every SEO tool you have tried ended the same way: a list of two hundred findings, sorted by severity, and no time to act on any of them. This agent crawls your site as a weekly Loop, decides which findings are worth the week, and then opens a pull request that fixes them on the repository that serves the site. You read the diff and merge it.

GitHub

The audit told you what was wrong and then left

You ran an SEO audit once. It gave you 214 findings, sorted by a severity score somebody else chose, and you fixed four of them on a Sunday and never went back. The site is still missing the canonical on half its pages, the sitemap still lists two URLs that were deleted in June, and you still cannot tell which of those things is costing you anything.

What is technical SEO?

Technical SEO is everything about a site that decides whether a search engine can read it properly: the titles and the headings, the canonical on every page, whether the sitemap matches what is actually served, whether robots.txt is letting the right crawlers in, whether the structured data says what the page says, whether the internal links reach every page you care about. None of it is writing. All of it is the reason a page that deserves to rank does not.

An SEO agent is a crawler that finishes the job. Instead of handing you a list of two hundred warnings, it crawls the site, compares it to last week, decides which findings are worth your week, and then opens a pull request that fixes them in your own repository. The audit is the boring half. The fix is the half nobody else does, because it needs to touch your code rather than a dashboard.

How it works

  1. 1

    Connect GitHub

    Connect the repository that serves your site and the agent rebuilds the picture before each audit: a crawl of the site covering titles, headings, canonicals, structured data, links and the site files, read against Search Console, the link profile and the crawl before it.How the GitHub connector works

  2. 2

    Runs weekly

    Once a week. The audit run reads the fresh crawl and writes down what is worth fixing. The Fix Runs that follow it open the pull requests, up to five, on the most urgent findings that do not have one yet.

  3. 3

    You read what it found

    There is nothing to approve in the feed, because a site fix is not a proposal. It is a recommendation, and its fix is a draft pull request on its own branch against the repository that serves your site. You read the diff. Approve merges it, Dismiss closes it and retires the recommendation, and nothing reaches your default branch any other way. A merge the repository refuses comes back as failed, with the pull request there to finish by hand.

What it leaves you

No proposals, because a site fix never goes through the approval feed. This agent leaves recommendations, and the Fix Runs after each audit turn the most urgent ones into pull requests on your repository, five a week at most.

The skills it runs

A run may activate these and nothing else. Every one of them is open: read the file before you turn the agent on.

Am I letting an agent commit to my repository?

Only as a pull request

Only as a pull request, on a branch of its own, one fix per branch. Kwern has no path to your default branch except the merge you ask for by approving the recommendation.

One finding per pull request

Each pull request is a single finding, so the diff is small enough to read properly rather than a hundred-file refactor you have to trust.

Five a week at most

There are at most five a week, on the findings that have no pull request yet. The queue is designed to be one you finish, not one you declare bankruptcy on.

What the AI CMO market sells as SEO

These come from those companies' own pricing pages and sitemaps on the day shown. The last one is the reason this agent exists: a site whose every URL points its canonical at the home page has a technical fault no amount of writing fixes. Kwern publishes no ranking or traffic figure of its own, because we have not measured one.

$129 a month

Okara's cheapest paid plan, which is where its AI SEO Agent starts

okara.ai/pricing, read 2026-09-12

$99 a month

RankSpot's single plan, which its own page lists against $149

www.rankspot.ai, read 2026-09-12

222 of 239

Hoox URLs that are blog posts, every one of them canonicalised to the home page

tryhoox.com/sitemap.xml, read 2026-09-12

Kwern against the alternatives

What is comparedKwernAI CMO suitesPoint toolsAn agencyDoing it yourself
Touching the site without breaking itA pull request on its own branch, one fix per branch, never a push to the branch you deploy from.A dashboard of findings, or a page published into a CMS you gave them the keys to.A crawler tells you. Changing anything is still your evening.A document of recommendations, and an invoice for writing it.You edit the template at midnight and find out on Thursday.
Finding what is brokenA fresh crawl of the site before each audit, read against Search Console, the link profile and the crawl before it, so you see what moved as well as what is wrong.A crawl, usually monthly, usually of the home page.A weekly crawl, with the same hundred warnings every week.A crawl at the start of the engagement, and not again.Search Console, when something has already gone wrong.
Writing the fixThe actual diff: the canonical, the heading, the schema block or the redirect, written into your files in your own conventions.A suggestion in prose, which you then implement.Nothing. A report is not a fix.A ticket for your developer, which is you.A search, a Stack Overflow answer and a hope.
Who says yes before it goes outOne feed. Approve, edit or reject each draft; nothing is published or spent until you do.Usually a dashboard, sometimes autopilot with no stop.Per tool, in whatever queue that tool keeps.A shared document, and a reply a day later.You wrote it, so you approved it.
What it costs a monthFrom $29 a month, every agent in every plan.$20 to $1,000 a month, and $5,000 for a done-for-you tier.One subscription per channel, each with its own seat.A monthly retainer, agreed before anything ships.Nothing, and it costs you the evenings.
Who pays for the modelYour own key. Across the 11 models Kwern vets, input runs $0.10 to $10 per million tokens, billed to you by the provider.Credits you buy from them; the markup is never itemised.Theirs, folded into the subscription.Inside the retainer, at whatever margin they keep.Your own key, at the provider's price.
Kwern and the four other ways of doing this work, on 6 rows. The suite prices were read from those companies' own pricing pages on 12 September 2026.

RiseMore, the cheapest

www.risemore.app/pricing

MarketOwl, the dearest

marketowl.ai/pricing

This Agent runs about 4 times a month. The tokens those runs spend are billed to you by your own provider: $0.10 per million in on gpt-6-luna, $10 on gpt-6-astra. Kwern never resells a token and never marks one up.

Three plans, every agent in each

Your model key, your accounts, your approval on every post. The plan decides how many products you run.

MonthlyYearly-17%

Lite

/ month

Billed monthly

Includes:

  • 1 Product
  • Usage included
  • Every agent, every skill and every service you can connect

Pro

/ month

Billed monthly

Includes:

  • 3 Products
  • 3× Lite's usage
  • Every agent, every skill and every service you can connect

Max

/ month

Billed monthly

Includes:

  • 5 Products
  • 5× Lite's usage
  • Every agent, every skill and every service you can connect
What the subscription covers

Questions about the SEO Agent

Someone is asking for what you built today.

On Reddit, on X, in ChatGPT's answers. Kwern finds them every morning and drafts the reply in your voice; you decide what goes out.