<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Now — remake</title>
    <link>https://www.tryremake.com/now</link>
    <atom:link href="https://www.tryremake.com/now/feed.xml" rel="self" type="application/rss+xml"/>
    <description>Product updates and articles from remake.</description>
    <language>en</language>
    <lastBuildDate>Fri, 20 Nov 2026 12:00:00 +0000</lastBuildDate>
    <item>
      <title>We’re social!</title>
      <link>https://www.tryremake.com/now/remake-on-x-and-linkedin</link>
      <guid isPermaLink="true">https://www.tryremake.com/now/remake-on-x-and-linkedin</guid>
      <pubDate>Fri, 20 Nov 2026 12:00:00 +0000</pubDate>
      <category>Update</category>
      <description>Follow remake on X and LinkedIn.</description>
      <content:encoded><![CDATA[<p>Follow remake on <a href="https://x.com/tryremakeai">X</a> and <a href="https://www.linkedin.com/company/145488110/">LinkedIn</a>.</p>
<p>#remake #AIVisibility #GEO #AISearch</p>]]></content:encoded>
    </item>
    <item>
      <title>Sign up with ChatGPT or Claude, then let your AI do the work</title>
      <link>https://www.tryremake.com/now/2026-11-16-sign-up-with-your-ai</link>
      <guid isPermaLink="true">https://www.tryremake.com/now/2026-11-16-sign-up-with-your-ai</guid>
      <pubDate>Mon, 16 Nov 2026 12:00:00 +0000</pubDate>
      <category>Update</category>
      <description>Also new: your AI sets up brands and makes your fixes, free calls, the pages AI cites instead of you, one Wallet for every limit, and Managers for your team.</description>
      <content:encoded><![CDATA[<h2 id="sign-up-with-chatgpt-or-claude">Sign up with ChatGPT or Claude</h2>
<p>On the <a href="https://www.tryremake.com/start">start page</a>, pick <strong>ChatGPT</strong> or <strong>Claude</strong>, then <strong>Start with ChatGPT</strong> (or Claude). Add remake to your AI, and it sets up your first website with you. Then create your account and confirm your email. From then on, <strong>Start in ChatGPT</strong> or <strong>Start in Claude</strong> opens a new chat with remake.</p>
<h2 id="your-ai-sets-up-brands-and-makes-your-fixes">Your AI sets up brands and makes your fixes</h2>
<p>Connect ChatGPT, Claude or Cursor from the new <strong>Your AI</strong> page. Ask your AI to set up a brand, and it walks you through what&rsquo;s missing, one step at a time, with each step&rsquo;s price, and waits for your yes.</p>
<p>Your AI can work your to-do list too: tick steps, mark fixes done and add notes. When it records a change on your site, remake can read that page and tell you whether the change is live. After a few runs, it also says whether your answers moved since.</p>
<p>Every call your AI or your own code makes is free, reads and changes alike, up to 120 calls a minute across the whole account. Work a call starts, such as a run, still shows its price first.</p>
<h2 id="see-the-pages-ai-cites-instead-of-you">See the pages AI cites instead of you</h2>
<p>Open the Fix Workshop and click <strong>Outside pages</strong>. It lists the pages AI cited in answers that named a rival and not you, grouped by site, from your newest run. One click reads up to 10 of them and shows which ones name you. You pay only for pages remake can read.</p>
<h2 id="one-wallet-and-managers-who-run-the-day-to-day">One Wallet, and Managers who run the day-to-day</h2>
<p>The Wallet now has four tabs: <strong>Overview</strong>, <strong>Limits</strong>, <strong>Spending</strong> and <strong>History</strong>. As the account owner, <strong>Limits</strong> lists your AI&rsquo;s limit, each Manager&rsquo;s and each brand&rsquo;s monthly cap, with this month&rsquo;s spend beside each. <strong>Spending</strong> splits a period by category, brand and person, and <strong>History</strong> shows one line for each run on each day.</p>
<p>On <strong>Team</strong>, invite a colleague as a <strong>Manager</strong>. Managers run the day-to-day work on your brands from your wallet, each under their own monthly limit; top-ups, your card and the team stay yours. Owners and Managers can also <strong>Archive</strong> a brand: its runs and reports stay, its schedule pauses, and it leaves your lists until you unarchive it.</p><h2>Improvements</h2><ul><li><strong>Search demand</strong> on Prompts scores each prompt from 1 to 5 by how often people search for it on Google each month, and <strong>Most searched</strong> sorts by it.</li><li>A Quick start and a lesson for each page show you around the app. Replay any of them from the Help menu under <strong>Lessons</strong>.</li><li>On the <strong>Rivals</strong> tab, <strong>Suggest</strong> offers up to 8 real companies with their websites, starting with rivals your runs already named.</li><li>In Insights, your chosen rivals lead the rival menu. <strong>Add</strong> any other name the answers carried, or hide it with <strong>Not a rival</strong>.</li><li>Pick <strong>Global (no country)</strong> for a brand whose customers aren&rsquo;t in one country. Its prompts name no place.</li><li>Six free website checkers, from AI crawlers to structured data, are open to anyone at <a href="https://www.tryremake.com/tools">tryremake.com/tools</a>, no sign-up needed.</li></ul><h2>Fixes</h2><ul><li>A website that answers only at its www address is now read there.</li><li>A scheduled run that comes due while your account is busy now waits its turn, and the app shows it&rsquo;s waiting.</li></ul><p>Correction, 18 Nov 2026: The <strong>Suggest</strong> button has moved out of the <strong>Rivals</strong> tab. In <strong>Brand settings → Rivals</strong>, you now see businesses named in your answers. Add them to your list or choose <strong>Not a rival</strong> to hide them. Finding names from your runs is free.</p>]]></content:encoded>
    </item>
    <item>
      <title>Is your website ready for AI search? Only 6 of 1,000 brand websites passed every check</title>
      <link>https://www.tryremake.com/now/brand-websites-ai-readiness</link>
      <guid isPermaLink="true">https://www.tryremake.com/now/brand-websites-ai-readiness</guid>
      <pubDate>Thu, 15 Oct 2026 12:00:00 +0000</pubDate>
      <category>Article</category>
      <description>Only 6 of 1,000 brand websites passed every check. The 12 basics that help AI read a site, how many brands have each, and the FAQ most of them miss.</description>
      <content:encoded><![CDATA[<p>It&rsquo;s Monday, and a screenshot lands in your inbox (from a client, or from your boss). Someone asked an AI assistant about your category, and a rival came back instead. Before anyone rewrites the brand story, ask a duller question. Can an assistant read your website, and does it answer what buyers ask?</p>
<p>AI assistants read plenty of pages about your brand. Your own website is the only one you fully control. So we read 1,000 brand websites and ran each through our website check. It looks for 12 basics that help AI reach and read a site, plus some smaller extras. Only six passed every check without a single warning, and they come from all over the map: a campsite and a print shop, a software company and an agency, an online doctor service and a group of care homes. (Whatever you sell, your site can join them.)</p><ul><li>Only <strong>6 of 1,000</strong> brand websites passed every check with no warnings. Camping Armanello, WIRmachenDRUCK and Benevity are among them.</li><li>All 12 basics are in place on <strong>157 of the 1,000 (16%)</strong>. All but six still carry a warning (our note on something that could be better).</li><li>The FAQ is the basic brands miss most. We found none on <strong>585 of the 1,000 (58%)</strong>, and for 122 sites it&rsquo;s the only basic missing.</li><li>Most finance and insurance brands have an FAQ (<strong>60 of 91,</strong> or 66%). Most food and drink brands don&rsquo;t (only 24 of 91 do, or 26%).</li></ul><p><strong>1,000 brand websites in our website check</strong></p><table><tbody><tr><td>Passed every check</td><td>6 (1%)</td></tr><tr><td>All 12 basics, with warnings</td><td>151 (15%)</td></tr><tr><td>Missing only an FAQ</td><td>122 (12%)</td></tr><tr><td>The rest: another basic missing</td><td>721 (72%)</td></tr></tbody></table><p>Each square is one site. The 6 and the 151 in pale teal have all 12 basics; the 122 in teal miss only an FAQ. Source: remake study of 1,000 brand websites read 14–15 Oct 2026. Company list from Wikidata (CC0).</p>
<h2 id="how-to-be-seen-for-ai-search">How to be seen for AI search?</h2>
<p>Exactly what a visitor needs (AI assistants included). The oldest trick in the world! Who you are, what you do, and how to reach you, all in plain text.</p>
<p>Run your eye down the bars. The technical basics hold up well (at least 924 sites on each of the first seven bars). Then the numbers thin out. Fewer than half have code saying who runs the site, and fewer still have an FAQ.</p>
<p><strong>The 12 basics, and how many of the 1,000 brand websites have each</strong></p><table><tbody><tr><td>Text that shows without running scripts</td><td>990</td></tr><tr><td>AI crawlers free to read the site</td><td>979</td></tr><tr><td>A homepage that loads without a long wait</td><td>972</td></tr><tr><td>A page title</td><td>966</td></tr><tr><td>A secure connection</td><td>962</td></tr><tr><td>A layout set up for phones</td><td>953</td></tr><tr><td>At least a few hundred words of text</td><td>924</td></tr><tr><td>A phone number or email address</td><td>824</td></tr><tr><td>A page description for search results</td><td>797</td></tr><tr><td>A main heading on the homepage</td><td>695</td></tr><tr><td>Code that says who runs the site</td><td>457</td></tr><tr><td>An FAQ page or section</td><td>415</td></tr></tbody></table><p>Each bar is out of the 1,000 brand websites we read in full. Source: remake study of 1,000 brand websites read 14–15 Oct 2026. Company list from Wikidata (CC0).</p>
<h2 id="what-do-the-best-brand-websites-do">What do the best brand websites do?</h2>
<p>We got to know three of them well. They have nothing in common, and that&rsquo;s the point. Each one does something any brand can copy.</p>
<ul>
<li><strong>Camping Armanello</strong>, a family campsite in Benidorm, wants you there in winter. It doesn&rsquo;t ask you to take its word for it. &ldquo;No lo decimos nosotros: lo dicen los datos,&rdquo; says the homepage (&ldquo;we&rsquo;re not saying it, the data is&rdquo;). Then come five winters of readings from AEMET, Spain&rsquo;s state weather agency. No significant rain on 89% of days from November to March, and an average high of 18.9°C. (Hard to argue with a weather station.)</li>
<li><strong>WIRmachenDRUCK</strong>, an online print shop in Backnang, Germany, leaves a business card in its homepage code. Visitors never see it. Software asking who runs the site finds the company&rsquo;s full name, its street address and a link to its Wikipedia article.</li>
<li><strong>Benevity</strong>, a Canadian company, makes workplace giving software. Its homepage opens with a slogan (&ldquo;Trust at scale. Undeniably proven.&rdquo;) and right after, it says what it sells. &ldquo;The Benevity® Enterprise Impact Platform is one trusted platform for employee giving, volunteering, grants and reporting.&rdquo; The slogan is for the heart. The sentence is for anyone who has to explain the company to someone else.</li>
</ul>
<p><a href="https://www.tryremake.com/#hero">Run the free website check on your own site</a></p>
<h2 id="do-you-need-an-faq-page-for-ai-search">Do you need an FAQ page for AI search?</h2>
<p>Before anyone buys, they have small, practical questions. What does it cost? When will it arrive? Are there nuts in it? Can I bring the dog? Now they can just ask an AI assistant. An FAQ is where you write the answers down, in your own words.</p>
<p>Where the answers sit matters too. Picture your delivery times folded away behind a little plus sign. A person clicks. Software might not. Microsoft&rsquo;s <a href="https://about.ads.microsoft.com/en/blog/post/october-2025/optimizing-your-content-for-inclusion-in-ai-search-answers">guide to appearing in AI answers</a> says it plainly: &ldquo;AI systems may not render hidden content.&rdquo; So keep every answer open on the page.</p>
<p>How much an FAQ sets you apart depends on your neighbours. For a bank or an insurer, it&rsquo;s keeping up. In food and drink, it puts you ahead of most of the category. A few to start with:</p>
<ul>
<li><strong>Food and drink:</strong> Is it gluten-free? Where can I buy it? How long does it keep once it&rsquo;s open?</li>
<li><strong>Finance and insurance:</strong> Can I apply? How long does approval take? How do I make a claim?</li>
<li><strong>Online shops:</strong> Do you deliver to me? When will it arrive? How do I send it back?</li>
<li><strong>Travel and hospitality:</strong> What time is check-in? Is there parking? How far is it to the centre?</li>
</ul>
<p><strong>Brand websites with an FAQ</strong></p><table><tbody><tr><td>Finance and insurance brands</td><td>66% ±10</td></tr><tr><td>All 1,000 brand websites</td><td>42% ±3</td></tr><tr><td>Food and drink brands</td><td>26% ±9</td></tr></tbody></table><p>An FAQ = a page or section titled as one, or FAQ markup, on the pages we read. Whiskers show the margin of error. Source: remake study of 1,000 brand websites read 14–15 Oct 2026. Company list from Wikidata (CC0).</p>
<h2 id="how-to-add-faq-schema-markup">How to add FAQ schema markup</h2>
<p>Once the answers are up, the markup is a smaller job. It&rsquo;s a short block of code that tells software which line is the question and which is the answer. Bing&rsquo;s <a href="https://www.bing.com/webmasters/help/webmasters-guidelines-30fba23a">guidelines</a> (which also cover Copilot) say structured data &ldquo;may support clearer grounding&rdquo; (grounding is how Copilot picks the pages it answers from). A modest promise. Google <a href="https://developers.google.com/search/updates">stopped showing FAQ results</a> in Search on 7 May 2026, so older advice about winning them no longer applies.</p>
<p>Bing adds one rule. Markup &ldquo;must accurately reflect visible content.&rdquo; So the block repeats your FAQ word for word, one entry per question:</p>
<pre><code>{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [
    {
      "@type": "Question",
      "name": "Which countries do you deliver to?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "We deliver to the UK and every EU country, usually within five working days."
      }
    },
    {
      "@type": "Question",
      "name": "How do I return an order?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "Start a return from your account within 30 days, and we email you a free label."
      }
    }
  ]
}
</code></pre>
<p>Your web team adds it to the FAQ page&rsquo;s HTML inside this tag:</p>
<p><code>&lt;script type="application/ld+json"&gt;</code></p>
<p>Once it&rsquo;s live, give the check below your FAQ page&rsquo;s address. It shows whether every question has an answer.</p>
<p><a href="https://www.tryremake.com/tools/structured-data-check">Check your FAQ schema with the free structured data check</a></p>
<h2 id="what-else-can-a-brand-website-do-for-ai-search">What else can a brand website do for AI search?</h2>
<ul>
<li><strong>Put a sentence or two under your headline</strong>, as Benevity does. Only 237 of the 951 brand homepages we could score have one (25%), so it&rsquo;s a quick way to stand out. <a href="https://www.tryremake.com/now/brand-homepages-easy-to-quote">How to make your homepage easy for AI to quote</a></li>
<li><strong>Link to your Wikipedia article from your website&rsquo;s code</strong>, as WIRmachenDRUCK does. It&rsquo;s a small club. Just 50 of the 925 brands with a Wikipedia article have done it. <a href="https://www.tryremake.com/now/brand-websites-wikipedia-link">How to link your website to your Wikipedia article</a></li>
<li><strong>Give AI search crawlers their own entry in your robots file</strong> (the short note that tells visiting software what it may read). Already, 96% of the 1,124 brand websites that answered when we asked for the file let ChatGPT and Claude search in, and few name them. A line of their own keeps them in on the day someone tidies up the rules for everyone else. <a href="https://www.tryremake.com/now/ai-crawlers-brand-websites">How brand websites treat AI crawlers</a></li>
</ul>
<p>So, back to Monday and that screenshot. Here&rsquo;s a line for the client (or the boss). &ldquo;Before we argue about what AI says about you, let&rsquo;s make sure it can read what you say about yourself.&rdquo; Then write down the five questions your buyers ask before they buy. Answer them on a page anyone can open. It makes a better Monday reply than a new brand story.</p>]]></content:encoded>
    </item>
    <item>
      <title>Robots.txt and AI crawlers: 96% of 1,124 brand websites let ChatGPT and Claude search in</title>
      <link>https://www.tryremake.com/now/ai-crawlers-brand-websites</link>
      <guid isPermaLink="true">https://www.tryremake.com/now/ai-crawlers-brand-websites</guid>
      <pubDate>Thu, 15 Oct 2026 12:00:00 +0000</pubDate>
      <category>Article</category>
      <description>96% of 1,124 brand websites let ChatGPT and Claude search in. What each AI crawler does, and three robots.txt blocks ready to paste.</description>
      <content:encoded><![CDATA[<p>It&rsquo;s Thursday, and the web team has had enough of bots. Someone writes two lines that close the whole site to every visitor (<code>User-agent: *</code>, then <code>Disallow: /</code>). Nobody wants to vanish from Google, so they add a group that lets Googlebot back in by name. Coffee. Job done. ChatGPT search is now standing outside, along with every other AI search crawler, because nobody wrote their names down.</p>
<p>On 14 and 15 October we went looking for that file on brand websites, as part of <a href="https://www.tryremake.com/now/ai-readiness-study">The AI Readiness Study</a>.</p><ul><li><strong>96%</strong> of brand websites let ChatGPT search and Claude search in. That&rsquo;s 1,083 of the 1,124 that answered when we asked for their robots.txt.</li><li>Only <strong>32 of the 1,124 (3%)</strong> name either search crawler. Everywhere else, the rules for every visitor decide. Close a folder to everyone, and AI search is shut out of it too.</li><li><strong>51 of the 64</strong> sites with rules for GPTBot or ClaudeBot (OpenAI&rsquo;s and Anthropic&rsquo;s training crawlers) never name the same company&rsquo;s search crawler. Search runs on its own crawler, so the training rule never reaches it.</li><li>We sent a checker that announces itself to 1,287 live brand websites. At 246 of them (<strong>19%</strong>), it was refused, stalled or shown a bot check. Shops and online stores did this more often (32%) than finance and insurance (14%).</li></ul><h2 id="how-do-ai-crawlers-read-your-robotstxt">How do AI crawlers read your robots.txt?</h2>
<p>By name. ChatGPT search sends a crawler called OAI-SearchBot, and Claude search sends Claude-SearchBot. Each looks for a group with its own name. Finding none, it follows the <code>User-agent: *</code> lines, the rules for everyone.</p>
<p><strong>120 of the 1,124 brand websites name an AI crawler in their robots.txt</strong></p><table><tbody><tr><td>No robots.txt, or an empty one</td><td>98 (9%)</td></tr><tr><td>A file that names no AI crawler</td><td>883 (79%)</td></tr><tr><td>Names at least one AI crawler</td><td>120 (11%)</td></tr><tr><td>Answered with an error when asked for the file</td><td>23 (2%)</td></tr></tbody></table><p>1,124 brand websites that answered when we asked for their robots.txt, 14–15 Oct 2026. Source: remake study of brand websites drawn from Wikidata.</p>
<p>That&rsquo;s how brand sites let AI search in. By saying nothing at all. Fine, until those lines change (a relaunch, a tidy-up, a new agency). Naming GPTBot or ClaudeBot doesn&rsquo;t count, by the way. Those are training crawlers, and search has its own.</p>
<p><strong>51 of the 64 files with rules for GPTBot or ClaudeBot never name the same company&#x27;s search crawler</strong></p><table><tbody><tr><td>Leave the search crawler to the general rules</td><td>51 (80%)</td></tr><tr><td>Name the search crawler too</td><td>13 (20%)</td></tr></tbody></table><p>Files that name GPTBot or ClaudeBot, out of the 120 brand robots.txt files that name at least one AI crawler. Source: remake study of brand websites drawn from Wikidata.</p>
<p>The Thursday file isn&rsquo;t made up. Eleven sites with no AI lines close the whole site to everyone, and five of them let Googlebot back in by name. At those five, ChatGPT&rsquo;s, Claude&rsquo;s and Perplexity&rsquo;s search crawlers stay outside. So does every other AI search crawler on our list.</p>
<p>The fix is a group of their own, like <a href="https://www.tryremake.com/now/ai-crawlers-brand-websites#2-ai-search-and-assistants-open-training-closed">choice 2 below</a>.</p>
<p><a href="https://www.tryremake.com/tools/ai-crawler-check">Check which AI crawlers your robots.txt lets in</a></p>
<h2 id="what-does-each-ai-crawler-do">What does each AI crawler do?</h2>
<p>Most AI companies send more than one crawler, each with its own job and a name only an engineer could love. Here they are, as each company listed them on 14 October 2026.</p>
<table>
<thead>
<tr>
<th>Job</th>
<th>Crawlers</th>
<th>What it does</th>
</tr>
</thead>
<tbody>
<tr>
<td>AI search</td>
<td><code>OAI-SearchBot</code> (ChatGPT), <code>Claude-SearchBot</code>, <code>PerplexityBot</code>, <code>Amzn-SearchBot</code>, <code>Meta-WebIndexer</code>, <code>MistralAI-Index</code>, <code>DuckAssistBot</code> (DuckDuckGo), <code>PetalBot</code> (Huawei)</td>
<td>Crawls pages so they can appear in AI answers. Google&rsquo;s AI Overviews follow Googlebot&rsquo;s rules, so they need no line of their own.</td>
</tr>
<tr>
<td>Assistants</td>
<td><code>ChatGPT-User</code>, <code>Claude-User</code>, <code>Perplexity-User</code>, <code>Amzn-User</code>, <code>Meta-ExternalFetcher</code>, <code>MistralAI-User</code></td>
<td>Fetch a page when a person asks about it. OpenAI, Perplexity, Amazon and Meta say robots.txt may not apply to theirs.</td>
</tr>
<tr>
<td>Training</td>
<td><code>GPTBot</code>, <code>ClaudeBot</code>, <code>MistralAI-Training</code>, <code>CCBot</code></td>
<td>Collects pages to train AI models. CCBot builds Common Crawl&rsquo;s public archive of the web, one of the most widely used sources of AI training data.</td>
</tr>
<tr>
<td>Training and other products</td>
<td><code>Amazonbot</code>, <code>Meta-ExternalAgent</code></td>
<td>May train models and also improve their makers&rsquo; other products.</td>
</tr>
<tr>
<td>Permission only</td>
<td><code>Google-Extended</code>, <code>Applebot-Extended</code></td>
<td>Never visit. They tell Google and Apple whether their AI models may use your pages.</td>
</tr>
</tbody>
</table>
<h2 id="how-to-allow-ai-crawlers-in-robotstxt">How to allow AI crawlers in robots.txt</h2>
<p>Three ways, from most open to most choosy, each ready to paste. First, three things to tidy.</p>
<ul>
<li><strong>Delete old AI lines first</strong>, including names that aren&rsquo;t in the table above. Two of them, anthropic-ai and Claude-Web, turn up in 32 of the 120 files that name an AI crawler. Anthropic&rsquo;s own crawler page lists neither. A crawler reads every group with its name, so an old line can clash with a new one.</li>
<li><strong>Copy your private areas into each named group.</strong> Checkout and account pages, for example. Palo Alto Networks does exactly this. A crawler with its own group skips your <code>User-agent: *</code> lines.</li>
<li><strong>Leave Googlebot, Bingbot and Applebot out of these blocks.</strong> They crawl for Google&rsquo;s, Bing&rsquo;s and Apple&rsquo;s main search (Google&rsquo;s AI Overviews included). A block for AI shouldn&rsquo;t change how ordinary search sees you.</li>
</ul>
<h3 id="1-everything-open">1. Everything open</h3>
<p>969 of the 1,124 sites already work like this, without a single AI line. Writing it down keeps every AI crawler in, whatever happens to the rest of the file. Vertiv even gives each crawler a name tag. &ldquo;# OAI-SearchBot — real-time retrieval for ChatGPT search and citations&rdquo;, reads one.</p>
<pre><code># Every AI crawler: welcome
User-agent: OAI-SearchBot
User-agent: ChatGPT-User
User-agent: Claude-SearchBot
User-agent: Claude-User
User-agent: PerplexityBot
User-agent: Perplexity-User
User-agent: Amzn-SearchBot
User-agent: Amzn-User
User-agent: Meta-WebIndexer
User-agent: Meta-ExternalFetcher
User-agent: MistralAI-Index
User-agent: MistralAI-User
User-agent: DuckAssistBot
User-agent: PetalBot
User-agent: GPTBot
User-agent: ClaudeBot
User-agent: Amazonbot
User-agent: Meta-ExternalAgent
User-agent: MistralAI-Training
User-agent: CCBot
User-agent: Google-Extended
User-agent: Applebot-Extended
Allow: /
# Your private areas, for example:
# Disallow: /checkout/
</code></pre>
<h3 id="2-ai-search-and-assistants-open-training-closed">2. AI search and assistants open, training closed</h3>
<p>Search and assistant crawlers get one group that lets them in. Training crawlers get one that keeps them out. Neither cares about your <code>User-agent: *</code> lines.</p>
<p>Decide on Google-Extended first. Google says it covers training Gemini models and the pages the Gemini app draws on for its answers. Google Search is unaffected (AI Overviews included). Want to stay in the Gemini app&rsquo;s answers? Delete the last group.</p>
<pre><code># AI search and assistants: welcome
User-agent: OAI-SearchBot
User-agent: ChatGPT-User
User-agent: Claude-SearchBot
User-agent: Claude-User
User-agent: PerplexityBot
User-agent: Perplexity-User
User-agent: Amzn-SearchBot
User-agent: Amzn-User
User-agent: Meta-WebIndexer
User-agent: Meta-ExternalFetcher
User-agent: MistralAI-Index
User-agent: MistralAI-User
User-agent: DuckAssistBot
User-agent: PetalBot
Allow: /
# Your private areas, for example:
# Disallow: /checkout/

# Training: closed
User-agent: GPTBot
User-agent: ClaudeBot
User-agent: Amazonbot
User-agent: Meta-ExternalAgent
User-agent: MistralAI-Training
User-agent: Applebot-Extended
User-agent: CCBot
Disallow: /

# Gemini training and Gemini app answers: closed
User-agent: Google-Extended
Disallow: /
</code></pre>
<h3 id="3-close-particular-crawlers-by-name">3. Close particular crawlers by name</h3>
<p>Korn Ferry&rsquo;s file keeps out exactly one: Bytespider, run by TikTok&rsquo;s owner, ByteDance. Add a group like this to either choice, one <code>User-agent</code> line per crawler.</p>
<pre><code># Closed by name
User-agent: Bytespider
Disallow: /
</code></pre>
<p>PetalBot is the name brands close most often (in 34 of the 120 files, and for more than half of those it&rsquo;s the only one). Huawei says PetalBot builds the index for Petal Search and for recommendations in Huawei Assistant and AI Search. Close it, and your site is out of both.</p>
<h2 id="does-your-bot-protection-let-crawlers-in">Does your bot protection let crawlers in?</h2>
<p>Our checker gives its name at every door. It was turned away at 246 live sites. At 197 of those, it couldn&rsquo;t even read the robots.txt on that visit. A crawler turned away at the door never gets to read the rules inside.</p>
<p><strong>How often brand websites turned away an automated visitor that gave its name, by country or group of countries</strong></p><table><tbody><tr><td>Germany</td><td>7% ±3</td></tr><tr><td>France, Italy and Spain</td><td>17% ±5</td></tr><tr><td>Canada and Australia</td><td>22% ±6</td></tr><tr><td>US</td><td>24% ±4</td></tr><tr><td>UK</td><td>25% ±6</td></tr><tr><td>All eight countries</td><td>19% ±2</td></tr></tbody></table><p>1,287 live brand websites, 14–15 Oct 2026. The visitor gave its name; one visit per site, from one location. Source: remake study of brand websites drawn from Wikidata.</p>
<p>Bot protection is the likely cause, so start there. It can be set to recognise AI crawlers, because OpenAI, Anthropic and Perplexity publish the addresses their crawlers come from. Take three requests to whoever runs it.</p>
<ul>
<li><strong>Let in the crawlers your robots.txt welcomes.</strong> For OpenAI, Anthropic and Perplexity, match each one to the addresses they publish.</li>
<li><strong>Keep robots.txt open to every visitor</strong>, so every crawler can read your rules.</li>
<li><strong>Ask to see it working.</strong> Its own records should show those crawlers getting through.</li>
</ul>
<p>Back to Thursday, then. Keep the two lines that close the site, and keep Googlebot&rsquo;s group. Paste choice 2 underneath, with your private areas copied in. Same coffee, three more groups (two, if you keep the Gemini app), and ChatGPT search is back inside.</p>
<p class="now-again"><a href="https://www.tryremake.com/tools/ai-crawler-check">Check your live robots.txt after the change</a></p>]]></content:encoded>
    </item>
    <item>
      <title>Homepages and AI search: only 25% put a sentence under the headline</title>
      <link>https://www.tryremake.com/now/brand-homepages-easy-to-quote</link>
      <guid isPermaLink="true">https://www.tryremake.com/now/brand-homepages-easy-to-quote</guid>
      <pubDate>Thu, 15 Oct 2026 12:00:00 +0000</pubDate>
      <category>Article</category>
      <description>Only 25% of 951 brand homepages put a sentence under the headline. What the leaders write there, and a nine-line brief for AI answers.</description>
      <content:encoded><![CDATA[<p>Somewhere in your company, a homepage headline was signed off after three rounds of edits and one long meeting. It probably sounds great. Now read it out loud to someone who has never heard of you, and ask them what you sell.</p>
<p>An AI assistant reading your homepage sits in that stranger&rsquo;s seat. It needs a plain sentence it can repeat. The best place for one is right under the headline.</p><ul><li>Only <strong>25% of the 951 brand homepages we checked (237)</strong> put a sentence or two straight under the headline.</li><li>At least <strong>61% (579)</strong> could add one. 270 have a headline with no sentence under it, and 309 have no proper headline.</li><li>The typical brand homepage does half of the nine things we checked, and <strong>21%</strong> do three-quarters or more.</li><li>In finance and insurance, <strong>43%</strong> of homepages do three-quarters or more. In automotive it&rsquo;s 10%.</li></ul><p>Benevity&rsquo;s homepage opens on a slogan, and the next line says what it sells:</p>
<blockquote class="now-shown">
<p>Trust at scale. Undeniably proven.</p>
<p>The Benevity® Enterprise Impact Platform is one trusted platform for employee giving, volunteering, grants and reporting.</p>
</blockquote>
<p><strong>Benevity&#x27;s homepage: a slogan, then a sentence that says what it sells</strong></p><table><tbody><tr><td>The headline</td><td>Trust at scale. Undeniably proven.</td></tr><tr><td>The line under it</td><td>The Benevity® Enterprise Impact Platform is one trusted platform for employee giving, volunteering, grants and reporting.</td></tr></tbody></table><p>The top of benevity.com&#x27;s homepage, read 14 Oct 2026. Source: benevity.com.</p>
<p>A visitor knows what Benevity does before scrolling, and an AI assistant has Benevity&rsquo;s own words to repeat. We checked 951 brand homepages for a sentence or two in that spot, and for eight other things that make a page easy to quote.</p>
<p><strong>What sits under the headline on 951 brand homepages</strong></p><table><tbody><tr><td>A sentence or two straight under the headline</td><td>237</td></tr><tr><td>A sentence only under a question heading lower down</td><td>57</td></tr><tr><td>A headline with no sentence under it</td><td>270</td></tr><tr><td>No proper headline</td><td>309</td></tr></tbody></table><p>Each bar is out of the 951 brand homepages. The rest put a shorter or longer line under the headline. Source: remake, 951 brand homepages read 14–15 Oct 2026. Company list from Wikidata (CC0).</p>
<h2 id="what-should-go-under-your-homepage-headline">What should go under your homepage headline?</h2>
<p>A sentence that says what you sell. Most homepages without one fall into two groups.</p>
<p>On 270, the headline is followed by something that isn&rsquo;t a sentence, often a second heading and sometimes a tagline, a row of buttons or a picture. The fix is small. If a line already sits under the headline, turn it into a full sentence of ordinary text. It can keep its look.</p>
<p>The other 309 have no proper headline. That takes one note to the web team:</p>
<blockquote>
<p>Please mark the big line at the top of our homepage as its main heading, the H1, and keep it the only one.</p>
</blockquote>
<p>Then the sentence goes under it.</p>
<p><a href="https://www.tryremake.com/tools/answer-readiness-check">Check what sits under your homepage&rsquo;s headline</a></p>
<h2 id="how-to-brief-a-homepage-for-ai-answers">How to brief a homepage for AI answers</h2>
<p>Here&rsquo;s the brief in nine lines, with the share of the 951 homepages already doing each.</p>
<table>
<thead>
<tr>
<th>Line in the brief</th>
<th>Already done by</th>
</tr>
</thead>
<tbody>
<tr>
<td>1. One headline, set as the page&rsquo;s main heading.</td>
<td>50%</td>
</tr>
<tr>
<td>2. A sentence or two (8 to 60 words) under the headline or a question heading.</td>
<td>33% in either place, 25% under the headline</td>
</tr>
<tr>
<td>3. At least one heading worded as a question.</td>
<td>23%</td>
</tr>
<tr>
<td>4. Two or more section headings.</td>
<td>70%</td>
</tr>
<tr>
<td>5. A list or a table.</td>
<td>44%</td>
</tr>
<tr>
<td>6. At least 300 words of main text.</td>
<td>60%</td>
</tr>
<tr>
<td>7. Short sentences and everyday words (a reading-ease score of 50 or more).</td>
<td>38% of English pages</td>
</tr>
<tr>
<td>8. A page title of 10 to 60 characters.</td>
<td>74%</td>
</tr>
<tr>
<td>9. A description of 50 to 160 characters.</td>
<td>57%</td>
</tr>
</tbody>
</table>
<p>What the lines say is up to you. Our advice? Keep the slogan for the headline, let line 2 say what you do and for whom, and borrow your customers&rsquo; questions for the headings.</p>
<p>Lines 2, 3 and 7 are the three the fewest homepages manage. All three are writing, so start there. A copywriter can do them without a redesign (nobody has to move a pixel).</p>
<h2 id="what-do-the-best-brand-homepages-do">What do the best brand homepages do?</h2>
<p>Fourteen homepages passed the whole brief. Twelve of them ask a question in a heading and put a sentence straight under it. The best ask what a customer would ask, then answer it at once. Under &ldquo;What is HubSpot?&rdquo;, HubSpot says:</p>
<blockquote class="now-shown">
<p>HubSpot brings together your teams, tools, AI, and customer data on the same platform, so every customer interaction drives growth.</p>
</blockquote>
<p><strong>HubSpot&#x27;s homepage: a question heading, answered straight under it</strong></p><table><tbody><tr><td>A question heading</td><td>What is HubSpot?</td></tr><tr><td>The answer under it</td><td>HubSpot brings together your teams, tools, AI, and customer data on the same platform, so every customer interaction drives growth.</td></tr></tbody></table><p>From hubspot.com&#x27;s homepage, read 14 Oct 2026. Source: hubspot.com.</p>
<p>It&rsquo;s the first thing a new visitor wants to know, answered in plain sight.</p>
<p>The 14 also give an assistant more to quote, with twice the text of the typical homepage.</p>
<p><strong>Ten of the 14, by name:</strong> hubspot.com · benevity.com · bestinvest.co.uk · standardlife.co.uk · hebronbrick.com · rowa.de · sanitas.es · hrg-online.de · campingarmanello.com · prodinger.de</p>
<h2 id="why-do-finance-homepages-ask-questions-more-often">Why do finance homepages ask questions more often?</h2>
<p>Finance and insurance brands put a question in a heading more often than the average brand (38% of 88 homepages, against 23% of all 951). Standard Life asks, among others, &ldquo;How healthy are your finances?&rdquo; and &ldquo;How could your pensions add up?&rdquo; Each leads to a tool that works out the answer. Why finance? Our guess is that pensions, investments and cover raise questions long before anyone buys. (Nobody picks a pension on impulse.)</p>
<p><strong>Finance and insurance homepages against all 951</strong></p><table><tbody><tr><td>A heading worded as a question: Finance and insurance</td><td>38% of 88</td></tr><tr><td>A heading worded as a question: All 951</td><td>23% of 951</td></tr></tbody></table><p>88 finance and insurance homepages and all 951 brand homepages. Industry shares carry a margin of 6 to 11 points; the gap is wider than both margins added together. Source: remake, 951 brand homepages read 14–15 Oct 2026. Company list from Wikidata (CC0).</p>
<h2 id="how-to-test-your-homepage-headline">How to test your homepage headline</h2>
<p>Back to that stranger. Show them your headline and the line under it, and ask what you sell. If they answer in their own words, the line works. If they repeat the slogan back to you, start with line 2.</p>
<p>At the next meeting about the headline, don&rsquo;t fight for a new one. Keep the slogan and ask for a sentence under it. Benevity kept its slogan, after all. It just didn&rsquo;t leave it on its own.</p>
<p class="now-again"><a href="https://www.tryremake.com/tools/answer-readiness-check">Check your homepage&rsquo;s headline and the line under it</a></p>]]></content:encoded>
    </item>
    <item>
      <title>Only 50 of 925 brands link to their Wikipedia article in their website&#x27;s schema</title>
      <link>https://www.tryremake.com/now/brand-websites-wikipedia-link</link>
      <guid isPermaLink="true">https://www.tryremake.com/now/brand-websites-wikipedia-link</guid>
      <pubDate>Thu, 15 Oct 2026 12:00:00 +0000</pubDate>
      <category>Article</category>
      <description>Of 925 brand websites whose company has a Wikipedia article, only 50 link to it in their sameAs schema. Here&#x27;s the code to add yours.</description>
      <content:encoded><![CDATA[<p>There are two Hays. One is a recruitment company that finds people jobs. The other is a town in Kansas that once gave Wild Bill Hickok one (sheriff, briefly, in 1869). Say &ldquo;Hays&rdquo; and somebody has to work out which you mean.</p>
<p>The company settles it with a short list tucked into the code of hays.com, out of sight of every visitor, pointing to its Wikipedia article and its Wikidata entry (Wikipedia&rsquo;s sister database). Software asking who runs the site gets a straight answer. This Hays.</p>
<p><strong>One line in hays.com&#x27;s code says which Hays the site belongs to</strong></p><pre><code>&quot;sameAs&quot;: [
  &quot;https://en.wikipedia.org/wiki/Hays_plc&quot;,
  &quot;https://www.wikidata.org/wiki/Q324029&quot;,
  &quot;https://www.linkedin.com/company/hays&quot;
]</code></pre><table><tbody><tr><td>Hays plc (A recruitment company): The one hays.com&#x27;s code names</td><td>https://en.wikipedia.org/wiki/Hays_plc</td></tr><tr><td>Hays, Kansas (A town in Kansas)</td><td>https://en.wikipedia.org/wiki/Hays,_Kansas</td></tr></tbody></table><p>The list in the &quot;who runs the site&quot; block of hays.com&#x27;s homepage, in full. Source: hays.com and Wikipedia, read 14 Oct 2026.</p>
<p>Plenty of brands share a name (with a town, a band, a word in the dictionary). If yours has a Wikipedia article, the hard part is done. Other people wrote it. You just have to point at it.</p>
<p>On 14 and 15 October we read the websites of 925 brands with a Wikipedia article, and 50 of them point to it the way Hays does.</p><ul><li>Just <strong>50 of the 925</strong> brands with a Wikipedia article (5%) link to it from their website&rsquo;s code.</li><li>Where a brand already keeps a list of profiles in that code (<strong>249 of the 925</strong>), Wikipedia is one line away for 199 of them (80%).</li><li>More than half have no such code yet (<strong>529 of the 925,</strong> or 57%). Two in three of those already link their social profiles on the page.</li></ul><h2 id="what-is-sameas-schema">What is sameAs schema?</h2>
<p>Schema is code a website keeps for software, not people. Microsoft&rsquo;s <a href="https://about.ads.microsoft.com/en/blog/post/october-2025/optimizing-your-content-for-inclusion-in-ai-search-answers">guide to AI search answers</a> says it helps &ldquo;search engines and AI systems understand your content.&rdquo;</p>
<p>One block of it says who runs the site, and inside sits a list called sameAs, the brand&rsquo;s other addresses around the web. <a href="https://schema.org/sameAs">Schema.org</a>, which writes the rules for this code, says each entry should be a page &ldquo;that unambiguously indicates the item&rsquo;s identity&rdquo;. Its first example is &ldquo;the item&rsquo;s Wikipedia page&rdquo;.</p>
<p>Unambiguously! That&rsquo;s the whole Hays problem in one word.</p>
<p><a href="https://www.tryremake.com/tools/trust-check">Check your site&rsquo;s trust and identity details</a></p>
<h2 id="how-to-add-wikipedia-to-your-sameas-list">How to add Wikipedia to your sameAs list</h2>
<p>View your homepage&rsquo;s source and search for &ldquo;sameAs&rdquo;. We found a list on 249 of the 925 brand websites, nearly all of them naming social accounts, and Wikipedia made it onto 50 (Wealthsimple, Target Australia and IONOS among them).</p>
<p>That leaves 199 where Wikipedia is one more address in a list the web team already looks after. Put your Wikidata entry beside it.</p>
<p><strong>199 of 249 brands with a profile list can add their Wikipedia article in one line</strong></p><table><tbody><tr><td>Keep a list that includes the Wikipedia article</td><td>50 (20%)</td></tr><tr><td>Room for one line: Wikipedia not on the list yet</td><td>199 (80%)</td></tr></tbody></table><p>Base: the 249 of 925 brands with a Wikipedia article whose &quot;who runs the site&quot; block keeps a profile list. Margin ±5 points. Counts only that list, not links shown on the page. Source: remake, brand websites read on 14–15 Oct 2026. Company list from Wikidata (CC0).</p>
<h2 id="how-to-add-organization-schema-with-sameas">How to add Organization schema with sameAs</h2>
<p>No block at all? You&rsquo;re in the majority (529 of the 925). Two in three of those brands already link their social profiles somewhere on the page, so the addresses are there, ready to copy. If you have the block but no list (147 brands do), you need only the sameAs lines.</p>
<p><strong>363 of the 529 with no block already link to their social profiles</strong></p><table><tbody><tr><td>Link to social profiles on the page</td><td>363 (69%)</td></tr><tr><td>No profile links found</td><td>152 (29%)</td></tr><tr><td>Couldn&#x27;t check profiles</td><td>14 (3%)</td></tr></tbody></table><p>Base: the 529 of 925 brands with a Wikipedia article whose site has no &quot;who runs the site&quot; block. Source: remake, brand websites read on 14–15 Oct 2026. Company list from Wikidata (CC0).</p>
<pre><code>{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "Your Brand",
  "url": "https://www.yourbrand.com/",
  "logo": "https://www.yourbrand.com/images/logo.png",
  "sameAs": [
    "https://en.wikipedia.org/wiki/Your_Brand",
    "https://www.wikidata.org/wiki/Q00000000",
    "https://www.linkedin.com/company/yourbrand"
  ]
}
</code></pre>
<p>It goes in the homepage&rsquo;s head, inside this tag:</p>
<p><code>&lt;script type="application/ld+json"&gt;</code></p>
<p>Keep the logo line. It&rsquo;s the one most often missing from half-finished blocks.</p>
<h2 id="how-to-find-your-wikidata-id">How to find your Wikidata ID</h2>
<p>The Wikipedia line is easy. Copy your article&rsquo;s address from the browser, in any language (a German-only article links de.wikipedia.org).</p>
<p>The Wikidata line takes one more click. On your Wikipedia article, open the Tools menu and click the Wikidata link (in English it says &ldquo;Wikidata item&rdquo;, and every language has one). Your Wikidata page opens. The end of its address is your number, a Q followed by digits. Hays, the company, is Q324029.</p>
<p><strong>Where your Wikidata ID is: the Tools menu on your Wikipedia article</strong></p><table><tbody><tr><td>Open the menu</td><td>Tools</td></tr><tr><td>Then click</td><td>Wikidata item</td></tr></tbody></table><p>Hays plc&#x27;s article on English Wikipedia. The Wikidata page that opens ends in its number, Q324029. Source: Wikipedia and Wikidata.</p>
<h2 id="why-your-brand-name-should-match-everywhere">Why your brand name should match everywhere</h2>
<p>Write the &ldquo;name&rdquo; line exactly as your brand appears everywhere else, on your homepage, your social profiles and your Wikipedia article, so nobody has to ask which Hays you are. A third of the 925 give their name only in the page title. This line in the head puts it where many apps look when someone shares a page:</p>
<p><code>&lt;meta property="og:site_name" content="Your Brand"&gt;</code></p>
<h2 id="why-link-your-website-to-wikipedia">Why link your website to Wikipedia?</h2>
<p>Here&rsquo;s why we like it. Your website is you describing yourself, and a Wikipedia article is other people describing you, edited in public, footnotes and all. (Your homepage calls you brilliant. Wikipedia sticks to the dates.) The sameAs line joins the two, so software can check one against the other.</p>
<p>So, which Hays? hays.com answers in its own code, with a link to an article it didn&rsquo;t write. The town in Kansas can sort itself out. Your brand&rsquo;s article is already written too. It&rsquo;s just waiting for the link.</p>
<p class="now-again"><a href="https://www.tryremake.com/tools/trust-check">See which trust details your site already shows</a></p>]]></content:encoded>
    </item>
    <item>
      <title>ChatGPT and Gemini answers now come from the real chat</title>
      <link>https://www.tryremake.com/now/2026-10-12-chatgpt-gemini-real-chat</link>
      <guid isPermaLink="true">https://www.tryremake.com/now/2026-10-12-chatgpt-gemini-real-chat</guid>
      <pubDate>Mon, 12 Oct 2026 12:00:00 +0000</pubDate>
      <category>Update</category>
      <description>Also new: fact-checking against your own facts, prompts and a model set picked as you add a brand, and bigger pitches for agencies, set up before they run.</description>
      <content:encoded><![CDATA[<h2 id="chatgpt-and-gemini-answers-come-from-the-real-chat">ChatGPT and Gemini answers come from the real chat</h2>
<p>Your brands&rsquo; runs now show the answer a customer sees on screen, with the pages it links to.</p>
<p>Answers from 8 October on come from the real chat, so a trend line for either model can step on that date.</p>
<h2 id="mistral-qwen-and-meta-ai-join-your-ai-models">Mistral, Qwen and Meta AI join your AI models</h2>
<p>Meta AI is part of every model set, Standard included. Mistral and Qwen come with the Pro and Hyper sets.</p>
<h2 id="check-ais-claims-against-your-own-facts">Check AI&rsquo;s claims against your own facts</h2>
<p>Sentiment is now <strong>Perception</strong>. Its <strong>Market</strong> tab adds how the answers describe your rivals, and your Tone by topic. On the new <strong>Fact-checking</strong> tab, type your facts once and each check marks the claims that contradict them.</p>
<h2 id="pick-your-prompts-and-model-set-when-you-add-a-brand">Pick your prompts and model set when you add a brand</h2>
<p>Adding a brand now has you pick your prompts and model set before the first run. We fill in the country you sell in when your homepage shows it. Then you add your prompts, written with AI or your own, choose your model set and start. Your Overview fills in while the first run works.</p>
<h2 id="set-up-a-pitch-before-it-runs">Set up a pitch before it runs</h2>
<p>On an agency account, you now set up each pitch in <strong>Prospecting</strong> before it runs, with the prospect&rsquo;s rivals, topics and prompts. A pitch now asks ChatGPT, Perplexity and Google AI Overviews up to 25 prompts, and ends with the top three website fixes. A pitch asks ChatGPT its own way, so its numbers can differ from your brands&rsquo; runs.</p>
<h2 id="see-which-ai-bots-your-site-lets-in">See which AI bots your site lets in</h2>
<p>Open <strong>My website → All site checks</strong> and click <strong>Crawler instructions</strong>. The list groups the AI bots by job: Search, Assistant, Training + answers, Training and Agent. Each shows whether it&rsquo;s allowed, blocked site-wide or blocked on some pages, next to the rule in your robots file that decides it.</p><h2>Improvements</h2><ul><li>My website lists the pages on your site that AI cited in the time you pick, with how often, by which AI models, and a Folders view.</li><li>Change a brand&rsquo;s country under Brand settings → Countries, and review its prompts reworded for the new country first.</li><li>Your fix suggestions use your site&rsquo;s own addresses, phone numbers and page list, and stay as they are until you click <strong>Refresh</strong>.</li><li><strong>Run new prompts only</strong> asks just your new prompts, right away.</li><li>Insights ranks a brand by Position once three or more answers place it.</li></ul><h2>Fixes</h2><ul><li>A robots file that keeps out only AI training bots now passes: it asks those bots to stay out, while AI search and assistant bots can still read your site.</li></ul>]]></content:encoded>
    </item>
    <item>
      <title>Your AI assistant now shows the price before it spends</title>
      <link>https://www.tryremake.com/now/2026-09-28-assistant-shows-the-price-first</link>
      <guid isPermaLink="true">https://www.tryremake.com/now/2026-09-28-assistant-shows-the-price-first</guid>
      <pubDate>Mon, 28 Sep 2026 12:00:00 +0000</pubDate>
      <category>Update</category>
      <description>Your connected assistant now shows the price before a run, a schedule or AI-written prompts, and waits for your yes. The app comes in dark, and Grok, DeepSeek and Kimi join your AI models.</description>
      <content:encoded><![CDATA[<h2 id="your-ai-assistant-asks-before-it-spends">Your AI assistant asks before it spends</h2>
<p>A connected assistant can now start or pause a brand&rsquo;s schedule, write prompts with AI, and rename or move brands, or bring back one you deleted. Runs, schedules and prompts written with AI come back with their price first and start only once you agree. Each yes pays for one action.</p>
<h2 id="dark-mode">Dark mode</h2>
<p>Pick <strong>Light</strong>, <strong>Dark</strong> or <strong>System</strong> under <strong>Appearance</strong>. <strong>System</strong> follows your computer&rsquo;s setting.</p>
<h2 id="grok-deepseek-and-kimi-join-your-ai-models">Grok, DeepSeek and Kimi join your AI models</h2>
<p>Choose the Hyper set to get all three.</p>
<h2 id="move-in-from-another-tool">Move in from another tool</h2>
<p>Upload the prompt export from your old AI-visibility tool, or let your AI assistant bring it over, rivals included. Your brands, prompts and workspaces arrive ready, and you start each brand&rsquo;s first run when you choose. Moving in is free.</p>
<h2 id="one-home-for-all-your-clients">One home for all your clients</h2>
<p>On an agency account, <strong>Workspaces</strong> now shows each client with its brands, this month&rsquo;s spend and the prompts it tracks. Open a row to see its brands, each with its score, spend and last run.</p><h2>Improvements</h2><ul><li>Schedule runs every hour, every 6 or 12 hours, daily, every 3 days, weekly or every 2 weeks.</li><li>Delete a brand and restore its setup within 7 days. Its runs and reports stay.</li><li>Agencies can export each client workspace&rsquo;s spend as a CSV file.</li></ul>]]></content:encoded>
    </item>
    <item>
      <title>Connect Claude, ChatGPT or Cursor to your account</title>
      <link>https://www.tryremake.com/now/2026-09-14-connect-your-ai-assistant</link>
      <guid isPermaLink="true">https://www.tryremake.com/now/2026-09-14-connect-your-ai-assistant</guid>
      <pubDate>Mon, 14 Sep 2026 12:00:00 +0000</pubDate>
      <category>Update</category>
      <description>Your AI assistant can now read your results, and any report can go out as a PDF or as a panel on a website.</description>
      <content:encoded><![CDATA[<h2 id="connect-your-ai-assistant">Connect your AI assistant</h2>
<p>Open <strong>Integrations</strong> and connect Claude, ChatGPT or Cursor from its own card. From inside the chat, your assistant can then read your brands, runs and results, and add or change your prompts.</p>
<h2 id="put-a-report-on-any-website">Put a report on any website</h2>
<p>Each report also has a compact panel you can put on your own site or a client&rsquo;s. The panel shows your score, Visibility by AI model and the trend, and carries your name when white label is on.</p>
<h2 id="send-your-results-to-zapier-make-and-slack">Send your results to Zapier, Make and Slack</h2>
<p>Connect Zapier or Make from <strong>Integrations</strong>, and a change in your numbers can add a sheet row, post a message or open a ticket. The Slack digest posts one summary a week, with a line for each brand with a schedule, up to 8 brands.</p>
<h2 id="give-your-team-their-own-sign-in">Give your team their own sign-in</h2>
<p>Open <strong>Team</strong>, type a colleague&rsquo;s email and send them the invite link. They get a Viewer seat, to read every brand, run and report.</p><h2>Improvements</h2><ul><li>Create API keys under Integrations → API &amp; MCP and read your results from your own code.</li><li>Download any report as a PDF.</li><li>A small snippet on your site counts the visits that AI answers send you.</li></ul>]]></content:encoded>
    </item>
  </channel>
</rss>
