NEXT.JS SEO KIT

See your link before you ship it.

Paste real content and it drafts a starting title and description locally — no AI call, just sentence scoring and cleanup. Then see a live Google search preview, an X/Twitter card, and a Facebook/OG card, with character counts against the ranges search engines actually truncate at. Copy the finished <meta> block when it looks right. Further down: the other SEO files a Next.js site needs, as code.

Local analysis + phrase composition, not an AI model call — drafts several options, cycle them with Shuffle.

Google search preview (approximate)

yoursite.com ›

Untitled page

No description provided.

Cut at about 600px for the title and 920px for the description (measured in Arial). Google changes these without notice, so treat the cut-off as a guide.

X / Twitter card

1200×630 image

yoursite.com

Facebook / OG card

1200×630 image

yoursite.com

Saved looks

Keep this setup under a name and come back to it. Saved on this device only.

Add a title or description, or press Load example, to see the code.

Check tags you already have

Paste the head of a live page and see what is missing, too long or repeated: the title, description, canonical link, viewport, robots and the social tags. It reads what you paste; it does not fetch your site.

Which AI crawlers can read your site?

Paste your robots.txt and see which of 13 known AI crawlers it blocks: OpenAI, Anthropic, Google, Perplexity, Apple, Meta and Common Crawl. It follows the same rules crawlers use: a crawler's own group wins over *, and the longest matching rule wins. The robots.ts tab above can write the opt-out for you.

With no robots.txt, every crawler below is allowed.

  • GPTBot · OpenAI · AI training

    Collects content that may be used to train OpenAI models.

    Allowed
  • OAI-SearchBot · OpenAI · AI search

    Surfaces sites in ChatGPT search results.

    Allowed
  • ChatGPT-User · OpenAI · Fetches when a person asks

    Visits a page when a person asks ChatGPT to. OpenAI says robots.txt rules may not apply.

    Allowed
  • ClaudeBot · Anthropic · AI training

    Collects web content that may contribute to training Claude.

    Allowed
  • Claude-SearchBot · Anthropic · AI search

    Improves the quality of Claude search answers.

    Allowed
  • Claude-User · Anthropic · Fetches when a person asks

    Visits a page when a person asks Claude a question.

    Allowed
  • PerplexityBot · Perplexity · AI search

    Surfaces sites in Perplexity search. Perplexity says it is not used for training.

    Allowed
  • Perplexity-User · Perplexity · Fetches when a person asks

    Visits a page when a person asks. Perplexity says it generally ignores robots.txt.

    Allowed
  • Google-Extended · Google · Training opt-out token

    A robots.txt control, not a separate crawler: whether Google may use your content for Gemini training and grounding. It does not affect Google Search.

    Allowed
  • Applebot-Extended · Apple · Training opt-out token

    A robots.txt control, not a crawler: whether Apple may use content Applebot crawled to train its models.

    Allowed
  • CCBot · Common Crawl · Open web archive

    Builds an open archive of the web that other people reuse.

    Allowed
  • meta-externalagent · Meta · AI training

    Crawls the web to train AI models or to index content for Meta products.

    Allowed
  • meta-externalfetcher · Meta · Fetches when a person asks

    Fetches links at a user’s request. Meta says it may bypass robots.txt.

    Allowed

The other SEO files for a Next.js site

FAQ schema, robots.ts, sitemap.ts and an opengraph-image.tsx for the App Router. Fill in a few fields and copy the file.

The same questions and answers must be visible on the page. Google now shows FAQ rich results only for a small set of well-known government and health sites, so this will not give you the expandable result in search. It is still valid markup that other search and AI tools can read.

Add your questions, or press Try an example, to see the code.

Why the character counts matter

Google truncates title tags around 50–60 characters and meta descriptions around 130–160 — go past that and search engines cut your copy off mid-sentence or rewrite it entirely. These are the same ranges this site's own pages are built against, so the counters here aren't arbitrary — they're the actual thresholds this tool's own buildMetadata() helper is written to respect.

What "draft from your content" actually does

It's real, but it's not AI. Pasted text is split into sentences, each sentence is scored by how much its words overlap with the rest of the text (extractive summarization — the same family of technique search engines have used for snippet generation for years), and the highest-scoring sentence becomes the description. A short list of common filler phrases ("in today's fast-paced world", "leverage", "utilize") gets cleaned up along the way. There's no model call, nothing leaves your browser, and it won't invent facts that aren't in what you pasted — because it isn't writing anything, just picking and trimming what's already there.

FAQ

Next.js SEO kit questions

What it generates, what it does not check, and the one Google change worth knowing.

8 questions

What does the kit generate?

Google, X and Facebook previews with a pixel-width cut-off, and code for HTML meta tags, Next.js metadata, JSON-LD, FAQ schema, robots.ts, sitemap.ts and an opengraph-image.tsx for the App Router.

Will FAQ schema give me expandable results in Google?

Probably not. Google now shows FAQ rich results only for a small set of well-known government and health sites. The markup is still valid and other search and AI tools can read it. The questions and answers must also be visible on the page.

Why does robots.ts leave out /_next/?

Google needs the scripts and styles under /_next/ to render your page. Blocking them can stop it seeing your content, so the generator never writes that rule.

Does it check my live page?

No. The head check reads tags you paste in. It does not fetch your site, and nothing you type is sent anywhere.

Will opting out of AI training hurt my Google ranking?

No. Google says its Google-Extended control does not affect inclusion in Google Search and is not a ranking signal. The opt-out also leaves the search crawlers of OpenAI, Anthropic and Perplexity allowed, so those products can still find and link to you.

Does the AI crawler check read my live robots.txt?

No. It reads the text you paste. Open yoursite.com/robots.txt, copy it in, and it shows which of 13 known crawlers are blocked. The list comes from each vendor’s own documentation and can go out of date.

Can robots.txt stop an assistant from reading my page?

Not always. Crawlers that collect content on their own are meant to follow it. A fetch that a person triggers by asking an assistant to open a page may ignore it: OpenAI, Perplexity and Meta say so for their user-triggered fetchers, while Anthropic says its bots honor robots.txt.

Is the title and description draft written by AI?

No. It scores sentences in the text you paste and picks and trims real ones. There is no model call, so it cannot invent facts.

Free tool by Adesh Shukla — everything runs client-side, nothing is sent anywhere.