See what search and AI crawlers can read of your archive, and what your rules allow.
Kanso samples your public pages, shows what your robots.txt does for nine named AI crawlers, and drafts long-form articles from Search Console queries. Your editors decide what is published, and Kanso never publishes.
Free to start · No credit card required. See how it works
At a glance
A large archive and a changing relationship with crawlers
A publisher has thousands of pages, and the audience reaches them through search and, increasingly, through AI systems that read the same pages.
Crawler rules are a decision, and easy to lose track of
Which AI crawlers are allowed is set in robots.txt, often long ago, and few people can say what it does today.
The archive is too big to check by hand
Templates repeat across thousands of pages, so a missing tag or a canonical problem repeats with them.
Search demand points at gaps
Queries where the site sits just off page one show topics readers want, but picking and briefing them takes editorial time.
From a rules file nobody reads to a clear report
Kanso reports what is in place and drafts from your search data. It makes no editorial or access decisions.
- Reading robots.txt by hand to see which crawlers are allowed
- Auditing templates page by page
- Choosing the next topic from a spreadsheet of queries
- Each of nine named AI crawlers shown as allowed or disallowed by your rules
- A sample of up to 25 pages checked, with the evidence for each finding
- An article drafted for a query your site already appears for
Site health, crawler rules and a first draft
SEO, GEO and Writer carry the work. Coding applies only to a static HTML site.
SEO Agent
Samples up to 25 public pages, such as the homepage, section pages and a few articles, and turns Search Console data into recommendations with evidence.
- It produces
- Which section templates miss basic tags, headings or canonical links.
- Where it fits
- Shows which templates need attention.
GEO Agent
Reports whether each of nine named AI crawlers is disallowed in your robots.txt, looks for llms.txt and structured data, and estimates citability.
- It produces
- A readiness analysis and a labelled estimate.
- Where it fits
- Makes your current crawler rules visible.
Writer Agent
Picks a Search Console query where the site is not on page one and drafts one long-form article in your house voice.
- It produces
- A Markdown draft for an editor.
- Where it fits
- A starting point, not a finished piece.
Coding Agent
For a supported title, description or canonical finding on a static HTML page in a GitHub repository, prepares a small change and opens a pull request once the owner confirms.
- It produces
- A diff preview and a pull request.
- Where it fits
- Only for a site built as static HTML files.
A review loop an editorial team can run
Five steps. Kanso reads and drafts, and editors decide.
- 01Kanso
Describe the publication through its site
Point Kanso at your site and it reads the homepage and linked same-site pages, then writes five company documents including a brand voice. It reads what the site says about itself, so correct it where it is wrong.
Sample the archive
The analysis covers up to 25 public pages, so it samples your templates, not the archive. A problem that repeats across a template shows up on the sample, and your team checks the rest.
Report on crawler rules
It reads your robots.txt and reports, for each of nine named AI crawlers, whether it is disallowed from your homepage, and flags a blanket disallow. It cannot know whether a block is deliberate, so what to do is your call.
Draft from search demand
From the last 30 days of queries it keeps those with more than 50 impressions where you are not on page one, picks the most searched, and drafts an article. It takes no brief and cannot be pointed at a story.
- 05You
Edit, verify and publish
An editor treats the draft as unverified until it has been fact-checked, rewrites it to house standards, and publishes it in your own tools.
What you receive
Findings and drafts wait in your Agents Feed.
Site findings
Findings name the page, the reasoning and the data they rest on, and can become tracked tasks.
A crawler-rules report
Allowed or disallowed for each named AI crawler, plus llms.txt and structured-data status.
An article draft
A long-form Markdown starting point for an editor, on a query the site already appears for.
Editorial judgement stays with editors
Kanso reports on your rules. It does not change them, and it does not decide what your outlet publishes.
Kanso does
- Reads your public pages and your search data
- Reports what your robots.txt does for named crawlers
- Drafts one article per run
You decide
- Which crawlers to allow or block
- Whether a draft is accurate, fair and worth publishing
- What is changed on the site, and by whom
Kanso never
- Publishes or schedules an article
- Edits your robots.txt or your site
- Fact-checks or edits to editorial standards
Where this fits
Situations, not results. They depend on your site and your queries.
A review of crawler access
Read the GEO report to see exactly what your rules allow today, before deciding what to change.
A template used across the archive
A missing description on a section page template shows on the sample, and you fix it once.
A topic readers keep searching
If Search Console shows it on page two, the Writer Agent can draft a starting point for an editor.
How Kanso approaches a publication's site
Each point is something the product does today.
It states what it sampled
The analysis covers up to 25 pages and says so, so you know the limits of the report.
Crawler rules are shown, not judged
It reports what your robots.txt does and leaves the decision to you.
Drafts start from reader demand
The topic is a query your site already appears for.
What Kanso does not do for publishers
A newsroom needs more than this. These are the limits.
It does not publish to a content system
There is no CMS integration and no scheduling of articles. Drafts are Markdown you publish yourself.
It does not audit the whole archive
The analysis samples up to 25 pages.
It does not fact-check or edit
A draft is a starting point. Treat it as unverified until your editors have checked it.
It does not see your audience
No readership, subscriber, revenue or ad data. It reads public pages and Search Console.
It does not measure AI answers
It reports your rules and estimates citability. It does not ask any AI product about you or track citations.
It does not edit framework sources
The Coding Agent edits a single static HTML file, and it stops when it cannot prove which file builds a page.
Related use cases
Get Found in AI Search
Check how readable your site is to AI systems, estimate how citable it looks, and turn the gaps into actions.
What the crawler report shows and what it cannot tell you.
Turn Search Console Data Into Content
Turn Search Console opportunities into long-form article drafts in your brand voice, for you to review and publish.
The Writer workflow in more detail.
Ship Technical SEO Fixes
Turn supported technical SEO findings into GitHub pull requests you review. Kanso never merges or deploys.
The pull-request workflow, for static HTML sites.
B2C
Social and content drafts for consumer brands that need to stay consistent across channels.
For consumer brands with their own content to keep consistent.
Questions about this use case
Will Kanso tell me which AI crawlers my site blocks?
Yes. It reads robots.txt and reports, for each of nine named crawlers, whether it is disallowed from your homepage. That checks the rule you published, not whether a crawler has visited.
Does it check my whole archive?
No. The analysis samples up to 25 public pages.
Can it write news or fact-check?
No. The Writer Agent drafts one long-form article per run from your Search Console data, with no brief. Treat a draft as unverified.
Does it publish articles for us?
No. It has no CMS integration, and it does not schedule publication.
Does Kanso change our robots.txt?
No. It reports what the rules do. You decide whether to change them.
Does it see our readers or revenue?
No. It reads your public pages and, if you connect it, your Search Console data.
See plans and pricing · All use cases
See what your site tells search and AI crawlers.
Enter your website. Kanso samples your pages, reports on crawler rules and drafts an article from your search data.
Get started free · No credit card required