It reads the served HTML
The page the server returns, not what a browser paints after running your scripts. That is the first thing a crawler sees.
Live verification · No account · No AI credits
Check the live website,
not your agent’s claim.
Paste a URL. TDKSEO reads the real live page, its robots.txt and its sitemap, then shows the earliest problem it can actually prove. You get the evidence, one next step, and a repair task you can hand to Codex, Claude Code, Lovable or any other agent. After you deploy, check the live site again.
Reads the public HTML, robots.txt and sitemap only. No account, no database, no model API, and nothing is stored by tdkseo.
Try:
What this checker actually does
Most SEO tools give you a list of warnings and a score. That is easy to generate and hard to act on. TDKSEO does something narrower and more useful: it reads your live site the way a crawler first sees it, finds the earliest problem that the evidence proves, and stops there.
It checks robots.txt and the sitemap
Crawl permission, site-wide blocks, XML validity, and whether the sitemap uses the same host as your live page.
It compares canonical against reality
Protocol, host and path. A canonical that disagrees with the live URL is a contradiction you can see, not a theory.
It refuses to guess
No search volume, no keyword difficulty, no index status, no traffic estimate. Those need data this checker does not have.
What it will never claim
TDKSEO only reports what it can point at. Everything below stays UNKNOWN until you connect a real data source, because a confident wrong answer is worse than an honest gap.
- Google indexing status. The checker can prove a page is allowed to be indexed; it cannot prove Google has indexed it.
- Search volume and keyword difficulty. No keyword data provider is connected.
- Whether you can beat the current results. That needs the actual SERP and the pages in it.
- Backlinks and domain authority. No backlink index is connected.
- Visibility inside AI answers. Brand mentions and citations are not measured here.
- Real-user Core Web Vitals. Field performance data needs CrUX or Search Console evidence.
- Whether your content is better than the competition. Only a reader or a separate audit can judge that.
- How Google finally renders complex JavaScript. This checker never executes your page scripts.
How to use it
- Paste your production URL.Use the URL you want indexed, not a preview or staging deployment.
- Read the highest verified blocker.Evidence, why it matters, what the expected result is, and how to verify it.
- Copy the fix task into your agent.Codex, Claude Code, Lovable, or any agent that can edit and deploy your project.
- Deploy the change yourself.TDKSEO never touches your repository or your hosting.
- Press Verify Again.The same live checks run again. The blocker only clears when the live site proves it.
Where this sits in the bigger picture
This page is stage tooling for one specific workflow: you already shipped something, and you want to know what to fix next. The full sequence — live, crawlable, indexable, structured, demand, content, published, measured — is on the SEO roadmap. The roadmap explains what can be automated, what still needs outside evidence, and what comes after each stage.
Questions
Is it free?
Yes. There is no account, no trial, and no credit card. The checker runs on Cloudflare Workers and does not call a paid API.
Why did my site come back as “Unable to verify”?
Some sites refuse automated requests or return a bot-protection page. That is reported as unverified, not as an SEO failure.
Does it crawl my whole site?
No. It checks one URL, plus that site’s robots.txt and sitemap. A full-site crawler is deliberately out of scope.
Do you store the pages you fetch?
No. The response is parsed in memory for the checks and only structured findings are returned to your browser.