<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title><![CDATA[CiteProof]]></title><description><![CDATA[CiteProof]]></description><link>https://citeproof.hashnode.dev</link><image><url>https://cdn.hashnode.com/uploads/logos/6aed9e9a7c4ff1f15c9b088d/915890c0-3a53-4bf0-ae31-7bb7caf320d0.png</url><title>CiteProof</title><link>https://citeproof.hashnode.dev</link></image><generator>RSS for Node</generator><lastBuildDate>Wed, 18 Nov 2026 15:00:40 GMT</lastBuildDate><atom:link href="https://citeproof.hashnode.dev/rss.xml" rel="self" type="application/rss+xml"/><language><![CDATA[en]]></language><ttl>60</ttl><item><title><![CDATA[Organization JSON-LD from a brand fact sheet (template included)]]></title><description><![CDATA[A brand fact sheet template and the matching Organization JSON-LD, so AI engines find the same facts about your business everywhere.]]></description><link>https://citeproof.hashnode.dev/organization-json-ld-from-a-brand-fact-sheet-template-included</link><guid isPermaLink="true">https://citeproof.hashnode.dev/organization-json-ld-from-a-brand-fact-sheet-template-included</guid><category><![CDATA[SEO]]></category><category><![CDATA[Web Development]]></category><category><![CDATA[schema]]></category><category><![CDATA[AI]]></category><dc:creator><![CDATA[CiteProof]]></dc:creator><pubDate>Tue, 17 Nov 2026 14:39:51 GMT</pubDate><content:encoded><![CDATA[<p>AI assistants describe a business from facts they can find and cross-check: name, what it does, location, phone, founding year, people, official profiles. Those facts should be identical on your site, your listings and your structured data.</p>
<p>The simplest way to keep them identical is a fact sheet: one internal document with the official values that everything else copies from. Here is a template, and how to turn it into <code>Organization</code> structured data.</p>
<h2>The fact sheet</h2>
<table>
<thead>
<tr>
<th>Fact</th>
<th>Where it must match</th>
</tr>
</thead>
<tbody><tr>
<td>Exact business name</td>
<td>Header, footer, Google Business Profile, directories, social profiles</td>
</tr>
<tr>
<td>What you do (one sentence)</td>
<td>About page, meta description, listings</td>
</tr>
<tr>
<td>Location</td>
<td>Contact page, footer, listings</td>
</tr>
<tr>
<td>Phone</td>
<td>Contact page, footer, listings</td>
</tr>
<tr>
<td>Founding year</td>
<td>About page</td>
</tr>
<tr>
<td>People and roles</td>
<td>About page, team page, LinkedIn</td>
</tr>
<tr>
<td>Main services</td>
<td>Service pages, listings</td>
</tr>
<tr>
<td>Official profiles</td>
<td>Structured data (<code>sameAs</code>), footer links</td>
</tr>
</tbody></table>
<p>Keep a "last updated" date on it. When a fact changes, change it on the sheet first, then everywhere.</p>
<h2>The same facts as JSON-LD</h2>
<pre><code class="language-html">&lt;script type="application/ld+json"&gt;
{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "Example Business",
  "url": "https://www.example.com/",
  "logo": "https://www.example.com/logo.png",
  "description": "What you do, in one sentence.",
  "foundingDate": "2015",
  "telephone": "+1-555-555-0100",
  "address": {
    "@type": "PostalAddress",
    "streetAddress": "123 Example Street",
    "addressLocality": "Denver",
    "addressRegion": "CO",
    "postalCode": "80202",
    "addressCountry": "US"
  },
  "sameAs": [
    "https://www.linkedin.com/company/example",
    "https://www.facebook.com/example"
  ]
}
&lt;/script&gt;
</code></pre>
<p>Notes:</p>
<ul>
<li>Use <code>LocalBusiness</code> (or a more specific type) instead of <code>Organization</code> if customers visit a physical location.</li>
<li><code>sameAs</code> lists your official profiles. Only include profiles you control.</li>
<li>Every value should match the visible text on your site. Structured data that disagrees with the page is one more conflicting source.</li>
</ul>
<h2>Check it</h2>
<ol>
<li>Compare the sheet with five places: your site, Google Business Profile, social profiles and two directories.</li>
<li>Ask ChatGPT, Perplexity and Gemini "what do you know about [business name]?" and note any wrong fact and the source it cites.</li>
<li>Fix each error at its source.</li>
</ol>
<p>Full template: <a href="https://gist.github.com/citeproof/70d1a213bedfe069af884413e2869474">https://gist.github.com/citeproof/70d1a213bedfe069af884413e2869474</a></p>
<p>The method is lesson 4.2 of our free AEO and GEO course: <a href="https://citeproof.co/courses/aeo-geo/about-page-and-fact-sheet-for-ai?utm_source=hashnode&amp;utm_medium=parasite&amp;utm_campaign=fact-sheet">The About page and the fact sheet</a>.</p>
<p><em>I'm Adrian, founder of CiteProof.</em></p>
]]></content:encoded></item><item><title><![CDATA[A console snippet to check if your page answers first (for AI search)]]></title><description><![CDATA[A browser console snippet that lists every heading and the first 60 words under it, to check if your page answers first for AI search.]]></description><link>https://citeproof.hashnode.dev/a-console-snippet-to-check-if-your-page-answers-first-for-ai-search</link><guid isPermaLink="true">https://citeproof.hashnode.dev/a-console-snippet-to-check-if-your-page-answers-first-for-ai-search</guid><category><![CDATA[SEO]]></category><category><![CDATA[JavaScript]]></category><category><![CDATA[Web Development]]></category><category><![CDATA[AI]]></category><dc:creator><![CDATA[CiteProof]]></dc:creator><pubDate>Mon, 16 Nov 2026 15:53:20 GMT</pubDate><content:encoded><![CDATA[<p>AI search engines like ChatGPT, Perplexity and Google's AI Overviews don't cite whole pages: they pull out passages that answer a question and cite their source. A practical way to make those passages easy to take is to put the answer in the first lines of the page and in the first lines under every heading.</p>
<p>Checking that by eye on a long page is slow. Here is a small snippet you can paste in the browser console: it lists every H1–H3 and the first 60 words that follow it, so you can see at a glance which sections answer and which ones warm up.</p>
<h2>The snippet</h2>
<pre><code class="language-js">// Paste in the browser console on any page.
// Prints each heading with the first 60 words that follow it.
(() =&gt; {
  const words = (t) =&gt; t.replace(/\s+/g, ' ').trim().split(' ').filter(Boolean);
  const heads = [...document.querySelectorAll('h1, h2, h3')];
  const rows = heads.map((h) =&gt; {
    let text = '';
    let el = h.nextElementSibling;
    while (el &amp;&amp; !/^H[1-3]$/.test(el.tagName) &amp;&amp; words(text).length &lt; 60) {
      text += ' ' + el.innerText;
      el = el.nextElementSibling;
    }
    const w = words(text);
    return {
      level: h.tagName,
      heading: h.innerText.trim().slice(0, 80),
      first60: w.slice(0, 60).join(' '),
      words: w.length,
    };
  });
  console.table(rows);
  return rows.length + ' headings checked';
})();
</code></pre>
<p>It only reads siblings that follow each heading, so it works best on pages where headings and paragraphs sit at the same level (most blog and CMS templates). If your theme wraps every section in its own <code>&lt;div&gt;</code>, adapt the loop.</p>
<h2>How to read the table</h2>
<p>For each row, ask three questions:</p>
<ol>
<li><strong>Does the heading use the words of the question?</strong> "How much does a dental implant cost in Denver?" beats "Let's talk money".</li>
<li><strong>Do the first sentences give the answer?</strong> A number, a yes or no, a choice. Not "there are many factors to consider".</li>
<li><strong>Does the passage make sense on its own?</strong> A common guideline is that the first 40–60 words under a heading should answer it. That's a recommended practice, not a published engine rule, but it is the shape of a quotable passage.</li>
</ol>
<p>Rows with <code>words: 0</code> are headings with nothing under them before the next heading: worth a look too.</p>
<h2>Fixing a section</h2>
<p>The formula is two or three sentences, then the rest:</p>
<ul>
<li>name the topic with the words of the question;</li>
<li>give the answer;</li>
<li>then explain conditions, exceptions and examples.</li>
</ul>
<p>Before: "In the ever-evolving landscape of the dental industry, there are many factors to consider." After: "A dental implant in Denver usually costs between [low] and [high]. The price depends on three things."</p>
<p>(Prices left blank on purpose: it's an illustration.)</p>
<h2>Try it</h2>
<p>Run the snippet on one page tied to a question where AI answers don't cite you. Rewrite the first hundred words and three sections, then re-run it.</p>
<p>The method is lesson 5.1 of our free AEO and GEO course: <a href="https://citeproof.co/courses/aeo-geo/write-pages-ai-cites-answer-first?utm_source=hashnode&amp;utm_medium=parasite&amp;utm_campaign=answer-first">The answer in the first lines</a>.</p>
<p><em>I'm Adrian, founder of CiteProof.</em></p>
]]></content:encoded></item><item><title><![CDATA[robots.txt vs sitemap.xml vs llms.txt: what each file does for AI search]]></title><description><![CDATA[What robots.txt, sitemap.xml and llms.txt each do, which AI crawlers read them, and the setup mistakes that block AI search visibility.]]></description><link>https://citeproof.hashnode.dev/robots-txt-vs-sitemap-xml-vs-llms-txt-what-each-file-does-for-ai-search</link><guid isPermaLink="true">https://citeproof.hashnode.dev/robots-txt-vs-sitemap-xml-vs-llms-txt-what-each-file-does-for-ai-search</guid><category><![CDATA[AI]]></category><category><![CDATA[SEO]]></category><category><![CDATA[WordPress]]></category><category><![CDATA[webdev]]></category><dc:creator><![CDATA[CiteProof]]></dc:creator><pubDate>Fri, 06 Nov 2026 09:50:07 GMT</pubDate><content:encoded><![CDATA[<p>robots.txt, sitemap.xml and llms.txt all sit at the root of your site, and they're easy to mix up. In short: <strong>robots.txt</strong> tells crawlers what they may fetch, <strong>sitemap.xml</strong> lists the URLs you want discovered, and <strong>llms.txt</strong> is a short, curated Markdown summary meant for AI tools. They don't replace each other, and only the first two are established standards.</p>
<p>Here's what each one does, which bots actually care, and the mistakes I see most often when people set them up for AI search.</p>
<h2>The quick comparison</h2>
<table>
<thead>
<tr>
<th></th>
<th>robots.txt</th>
<th>sitemap.xml</th>
<th>llms.txt</th>
</tr>
</thead>
<tbody><tr>
<td>Purpose</td>
<td>Allow or block crawling of paths</td>
<td>List URLs for discovery</td>
<td>Curated summary of key pages for AI tools</td>
</tr>
<tr>
<td>Format</td>
<td>Plain-text directives</td>
<td>XML</td>
<td>Markdown</td>
</tr>
<tr>
<td>Status</td>
<td>Standard (RFC 9309)</td>
<td>Standard (sitemaps.org protocol)</td>
<td>Proposal (llmstxt.org)</td>
</tr>
<tr>
<td>Read by</td>
<td>Search and AI crawlers that respect it</td>
<td>Search engines</td>
<td>No confirmed use by major AI search engines</td>
</tr>
<tr>
<td>Controls indexing?</td>
<td>No, only crawling</td>
<td>No</td>
<td>No</td>
</tr>
</tbody></table>
<h2>robots.txt: access, not visibility</h2>
<p>robots.txt is a list of rules per user agent: which paths a crawler may or may not request. It's the one file of the three that AI companies explicitly document. OpenAI, Anthropic and Perplexity each publish the names of their crawlers (for example <code>GPTBot</code>, <code>OAI-SearchBot</code>, <code>ClaudeBot</code>, <code>PerplexityBot</code>) so site owners can allow or block them.</p>
<p>Two things people get wrong:</p>
<ol>
<li><strong>Blocking a path doesn't remove it from results.</strong> robots.txt controls crawling, not indexing. For that you need <code>noindex</code> or removal tools.</li>
<li><strong>Blocking AI crawlers by accident.</strong> Some security plugins and CDN "bot fight" settings block unknown bots by default. If you want to be cited, check that the crawlers you care about can fetch your pages.</li>
</ol>
<p>A minimal example:</p>
<pre><code class="language-txt">User-agent: *
Allow: /

User-agent: GPTBot
Allow: /

Sitemap: https://example.com/sitemap.xml
</code></pre>
<h2>sitemap.xml: discovery, not ranking</h2>
<p>A sitemap lists the canonical URLs you want crawled, optionally with <code>lastmod</code> dates. It helps crawlers find pages, especially new or deep ones. It doesn't make a page rank or get cited.</p>
<p>Common mistakes:</p>
<ul>
<li>Listing redirected, noindexed or 404 URLs</li>
<li><code>lastmod</code> dates that change on every build, which makes the field meaningless</li>
<li>Forgetting to reference it in robots.txt or submit it in Search Console</li>
</ul>
<h2>llms.txt: a curated summary</h2>
<p>llms.txt is a Markdown file at <code>/llms.txt</code> with a title, a one-paragraph summary, and grouped lists of your most important links, each with a short description. Think of it as the page you'd hand to someone who has five minutes to understand your site.</p>
<pre><code class="language-markdown"># Example Co

&gt; Example Co makes invoicing software for small agencies.

## Key pages
- [Pricing](https://example.com/pricing): three plans, monthly billing
- [How it works](https://example.com/how-it-works): setup in three steps
</code></pre>
<p>The honest caveat: there's no public confirmation that ChatGPT, Perplexity or Google AI Overviews use llms.txt when choosing what to cite. It's cheap to create and low-risk, so it's reasonable housekeeping. Just don't treat it as a lever on its own.</p>
<h2>Which one matters most for AI visibility?</h2>
<p>If you only fix one thing, fix robots.txt: an AI search crawler that can't fetch your pages can't cite them. Then keep the sitemap clean so new pages get found. Add llms.txt last.</p>
<p>None of the three decides whether an assistant recommends you. That depends on whether your pages answer the questions people ask, whether your structured data is clear, and whether other sources mention you.</p>
<h2>Checklist</h2>
<ul>
<li><input disabled type="checkbox" /> robots.txt allows the AI crawlers you want, and nothing important is blocked by a plugin or CDN rule</li>
<li><input disabled type="checkbox" /> sitemap.xml lists only live, canonical URLs and is referenced in robots.txt</li>
<li><input disabled type="checkbox" /> llms.txt (optional) lists your key pages with real descriptions, and every URL returns 200</li>
<li><input disabled type="checkbox" /> All three load as plain files at the root, not as HTML pages</li>
</ul>
<p>If you're on WordPress, I wrote a <a href="https://dev.to/citeproof/how-to-add-an-llmstxt-file-to-wordpress-and-what-it-will-and-wont-do-3gmj">step-by-step llms.txt walkthrough</a>. To generate a first draft from your own sitemap, you can use CiteProof's <a href="https://citeproof.co/tools/llms-txt-generator?utm_source=hashnode&amp;utm_medium=parasite&amp;utm_campaign=llms-txt-vs-robots">free llms.txt generator</a>.</p>
<p><em>Disclosure: I'm the founder of CiteProof.</em></p>
]]></content:encoded></item></channel></rss>