A plain-text version of this site for AI assistants is at https://contengi.com/llms.txt, with the full product documentation in one file at https://contengi.com/llms-full.txt. Every page is also available as markdown by adding .md to its URL.
A dark AI product site for an LLM evaluation platform, with a canvas orrery of test cases in the hero and a live table where you swap models and watch regressions appear.
AI product - LLM evaluation platform · 21 November 2026
Orrery is a one-page site for an AI developer tool that shows the product working instead of a glowing orb: an animated canvas of orbiting test cases, an interactive model swap that turns rows red, counters that run on scroll, a CI snippet, a security grid, and usage-based pricing.
Download the HTML, then paste it into Claude, ChatGPT, Cursor, or any coding agent with this prompt. Fill in the parts in brackets.
Here is a single-file HTML page called Orrery, a site for an AI evaluation platform. Keep the canvas hero with orbiting points that flip colour, the crosshair cursor with coordinates, the scramble effect on the sharp monospaced buttons, the interactive table where switching a toggle changes the scores and verdicts, the counters that run on scroll, and the code snippet beside the steps. Turn it into a site for [your AI product]: change the name and logo to [name], rewrite the headline for [the problem your product removes], change what the hero points represent to [your product's unit of work], rebuild the interactive table as a demo of [the one thing your product does best] with [realistic sample data], replace the counters with [your real numbers], rewrite the steps and snippet for [how a developer sets it up], and set pricing to [your plans]. Keep the near-black, cyan, and violet palette unless I give you brand colours. No network calls from the page. Respect prefers-reduced-motion. One HTML file only.
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>Orrery - Evals for teams shipping LLM features</title>
<meta name="description" content="Orrery runs your evals on every prompt, model, and tool change, and shows which customers would have seen the regression before you ship it.">
<link rel="preconnect" href="https://fonts.googleapis.com">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link href="https://fonts.googleapis.com/css2?family=Geist:wght@300;400;500;600;700&family=Geist+Mono:wght@400;500&display=swap" rel="stylesheet">
<style>
:root{--bg:#07090d;--panel:#10141b;--line:rgba(232,236,241,.1);--text:#e8ecf1;--dim:#8b95a5;--cyan:#5cf2d6;--violet:#9b8cff;--red:#ff6b6b}
*{box-sizing:border-box;margin:0;padding:0}
body{background:var(--bg);color:var(--text);font-family:Geist,system-ui,sans-serif;font-size:16.5px;line-height:1.6;-webkit-font-smoothing:antialiased;overflow-x:hidden}
a{color:inherit;text-decoration:none}
.mono{font-family:"Geist Mono",ui-monospace,monospace}
.wrap{max-width:1280px;margin:0 auto;padding:0 28px}
/* sharp buttons whose label scrambles on hover */
.btn{display:inline-flex;align-items:center;gap:10px;height:46px;padding:0 20px;border:1px solid var(--line);font-family:"Geist Mono",monospace;font-size:13.5px;letter-spacing:.02em;text-transform:uppercase;color:var(--text);background:transparent;cursor:pointer;transition:border-color .2s,background .2s}
.btn:hover{border-color:var(--cyan)}
.btn.solid{background:var(--cyan);color:#04110e;border-color:var(--cyan)}
.btn.solid:hover{background:#86f7e3}
.btn [data-scramble]{display:inline-block;white-space:nowrap}
/* crosshair cursor */
.xh{position:fixed;pointer-events:none;z-index:50;background:rgba(92,242,214,.35);opacity:0;transition:opacity .2s}
.xh.h{left:0;right:0;height:1px}
.xh.v{top:0;bottom:0;width:1px}
.xh-label{position:fixed;pointer-events:none;z-index:51;font-family:"Geist Mono",monospace;font-size:11px;color:var(--cyan);opacity:0;transition:opacity .2s;white-space:nowrap}
body.xh-on .xh,body.xh-on .xh-label{opacity:1}
@media (hover:none){.xh,.xh-label{display:none}}
nav{position:absolute;top:0;left:0;right:0;z-index:10}
nav .wrap{display:flex;align-items:center;justify-content:space-between;height:76px;gap:20px}
.logo{display:flex;align-items:center;gap:11px;font-weight:600;font-size:18px;letter-spacing:-.02em}
.logo svg{width:28px;height:28px}
.nl{display:flex;gap:28px;font-size:14px;color:var(--dim)}
.nl a:hover{color:var(--text)}
@media (max-width:860px){.nl{display:none}}
@media (max-width:460px){nav .btn{display:none}}
/* hero */
.hero{position:relative;min-height:100vh;display:flex;align-items:flex-end;padding:120px 0 64px;border-bottom:1px solid var(--line);overflow:hidden}
#sky{position:absolute;inset:0;width:100%;height:100%;display:block}
.hero:after{content:"";position:absolute;inset:0;background:radial-gradient(ellipse at 70% 40%,transparent 0,rgba(7,9,13,.2) 40%,var(--bg) 92%);pointer-events:none}
.hero .wrap{position:relative;z-index:2;width:100%}
h1{font-size:clamp(44px,7vw,108px);line-height:.95;letter-spacing:-.05em;font-weight:500;max-width:1000px}
h1 span{color:var(--cyan)}
.hero-row{margin-top:36px;display:grid;grid-template-columns:minmax(0,1fr);gap:28px;align-items:end}
@media (min-width:960px){.hero-row{grid-template-columns:minmax(0,1fr) auto}}
.hero p{font-size:19px;color:var(--dim);max-width:560px;font-weight:300}
.ctas{display:flex;flex-wrap:wrap;gap:10px}
.legend{margin-top:40px;display:flex;flex-wrap:wrap;gap:22px;font-size:12px;color:var(--dim)}
.legend i{display:inline-block;width:8px;height:8px;border-radius:50%;margin-right:8px;vertical-align:1px}
/* sections */
section{padding:120px 0;border-bottom:1px solid var(--line)}
h2{font-size:clamp(34px,4.4vw,62px);line-height:1;letter-spacing:-.045em;font-weight:500;max-width:860px}
.sub{margin-top:18px;color:var(--dim);font-size:18px;max-width:600px;font-weight:300}
/* the interactive regression check */
.lab{margin-top:56px;border:1px solid var(--line);background:var(--panel)}
.lab-top{display:flex;flex-wrap:wrap;gap:16px;justify-content:space-between;align-items:center;padding:16px 18px;border-bottom:1px solid var(--line)}
.seg{display:flex;border:1px solid var(--line)}
.seg button{background:none;border:0;color:var(--dim);font-family:"Geist Mono",monospace;font-size:12.5px;padding:9px 14px;cursor:pointer}
.seg button[aria-pressed=true]{background:var(--text);color:var(--bg)}
.score{display:flex;gap:26px;font-size:12.5px;color:var(--dim)}
.score b{display:block;font-size:26px;color:var(--text);font-weight:500;letter-spacing:-.03em}
.score b.bad{color:var(--red)}
.tbl{width:100%;border-collapse:collapse;font-size:14px}
.tbl th{font-family:"Geist Mono",monospace;font-weight:400;font-size:11.5px;color:var(--dim);text-align:left;padding:12px 18px;border-bottom:1px solid var(--line)}
.tbl td{padding:13px 18px;border-bottom:1px solid var(--line);vertical-align:top}
.tbl tr:last-child td{border-bottom:0}
.pill{font-family:"Geist Mono",monospace;font-size:11.5px;padding:3px 8px;border:1px solid currentColor;white-space:nowrap}
.pass{color:var(--cyan)}.fail{color:var(--red)}
.bar{height:6px;background:rgba(232,236,241,.08);margin-top:8px;width:120px}
.bar i{display:block;height:100%;background:var(--cyan);transition:width .5s}
.bar i.lo{background:var(--red)}
.tbl .who{color:var(--dim);font-size:12.5px}
@media (max-width:720px){.tbl .hide{display:none}.tbl td,.tbl th{padding:12px 12px}}
/* counters */
.counts{margin-top:64px;display:grid;grid-template-columns:repeat(auto-fit,minmax(230px,1fr));border-top:1px solid var(--line)}
.counts div{padding:28px 24px 0 0}
.counts b{display:block;font-size:clamp(46px,5.4vw,76px);font-weight:400;letter-spacing:-.05em;line-height:1}
.counts b small{font-size:.4em;color:var(--violet);letter-spacing:0;margin-left:4px}
.counts p{margin-top:12px;color:var(--dim);font-size:14.5px;max-width:250px}
/* steps + code */
.steps{margin-top:56px;display:grid;grid-template-columns:minmax(0,1fr);gap:28px}
@media (min-width:960px){.steps{grid-template-columns:minmax(0,.9fr) minmax(0,1.1fr)}}
.step{display:grid;grid-template-columns:44px 1fr;gap:16px;padding:22px 0;border-top:1px solid var(--line)}
.step span{font-family:"Geist Mono",monospace;color:var(--violet);font-size:13px}
.step h3{font-size:20px;font-weight:500;letter-spacing:-.02em}
.step p{color:var(--dim);margin-top:6px;font-size:15.5px}
pre{background:var(--panel);border:1px solid var(--line);padding:22px;font-family:"Geist Mono",monospace;font-size:13px;line-height:1.75;color:#c6cdd8;white-space:pre-wrap;word-break:break-word}
pre .k{color:var(--violet)}pre .s{color:var(--cyan)}pre .c{color:#5b6576}
/* trust */
.trust{display:grid;grid-template-columns:repeat(auto-fit,minmax(240px,1fr));gap:1px;background:var(--line);border:1px solid var(--line);margin-top:56px}
.trust div{background:var(--bg);padding:28px}
.trust h3{font-size:18px;font-weight:500}
.trust p{color:var(--dim);margin-top:8px;font-size:15px}
/* pricing */
.plans{margin-top:56px;display:grid;grid-template-columns:repeat(auto-fit,minmax(270px,1fr));gap:1px;background:var(--line);border:1px solid var(--line)}
.plan{background:var(--bg);padding:32px;display:flex;flex-direction:column;gap:18px}
.plan.hi{background:var(--panel);box-shadow:inset 0 2px 0 var(--cyan)}
.plan h3{font-family:"Geist Mono",monospace;font-weight:400;font-size:13px;text-transform:uppercase;letter-spacing:.06em;color:var(--dim)}
.plan .p{font-size:52px;letter-spacing:-.05em;line-height:1}
.plan .p small{font-size:15px;color:var(--dim);letter-spacing:0}
.plan ul{list-style:none;display:grid;gap:9px;font-size:15px;flex:1}
.plan li:before{content:"+ ";color:var(--cyan);font-family:"Geist Mono",monospace}
.end{text-align:center;padding:150px 0;border-bottom:0;position:relative;overflow:hidden}
.end h2{margin:0 auto}
.end .ctas{justify-content:center;margin-top:34px}
.end:before{content:"";position:absolute;left:50%;top:50%;width:900px;height:900px;transform:translate(-50%,-50%);border-radius:50%;border:1px solid var(--line);box-shadow:0 0 0 120px rgba(155,140,255,.025),0 0 0 240px rgba(92,242,214,.02)}
.end .wrap{position:relative}
footer{border-top:1px solid var(--line);padding:40px 0;font-size:13.5px;color:var(--dim)}
footer .wrap{display:flex;flex-wrap:wrap;justify-content:space-between;gap:16px}
footer nav{position:static;display:flex;gap:22px;flex-wrap:wrap}
.js .rv{opacity:0;transform:translateY(18px);transition:opacity .9s,transform .9s cubic-bezier(.2,.7,.2,1)}
.js .rv.in{opacity:1;transform:none}
@media (prefers-reduced-motion:reduce){*{transition:none!important;animation:none!important}.js .rv{opacity:1;transform:none}}
</style>
</head>
<body>
<div class="xh h" id="xhH"></div><div class="xh v" id="xhV"></div><div class="xh-label" id="xhL"></div>
<nav>
<div class="wrap">
<a href="#" class="logo"><svg viewBox="0 0 28 28" aria-hidden="true"><circle cx="14" cy="14" r="12" fill="none" stroke="#e8ecf1" stroke-opacity=".35"/><circle cx="14" cy="14" r="6.5" fill="none" stroke="#e8ecf1" stroke-opacity=".6"/><circle cx="14" cy="14" r="2.6" fill="#5cf2d6"/><circle cx="24.4" cy="8" r="2" fill="#9b8cff"/></svg>Orrery</a>
<div class="nl"><a href="#lab">Regressions</a><a href="#how">How it works</a><a href="#trust">Security</a><a href="#pricing">Pricing</a><a href="#">Docs</a></div>
<a href="#pricing" class="btn"><span data-scramble>Get an API key</span></a>
</div>
</nav>
<header class="hero">
<canvas id="sky" aria-hidden="true"></canvas>
<div class="wrap">
<h1>Ship the model update without the <span>2am rollback.</span></h1>
<div class="hero-row">
<p>Orrery runs your evals on every prompt, model, and tool change, then shows you which customers would have seen the regression while there is still time to fix it.</p>
<div class="ctas"><a href="#pricing" class="btn solid"><span data-scramble>Start free</span></a><a href="#lab" class="btn"><span data-scramble>Try a model swap</span></a></div>
</div>
<div class="legend mono"><span><i style="background:#5cf2d6"></i>Passing test case</span><span><i style="background:#ff6b6b"></i>Regression</span><span><i style="background:#9b8cff"></i>New coverage</span></div>
</div>
</header>
<section id="lab">
<div class="wrap">
<h2 class="rv">Swap the model. See what breaks.</h2>
<p class="sub rv">This is a real Orrery run on a support assistant. Switch between the current model and the candidate, and watch which cases regress and who they belong to.</p>
<div class="lab rv">
<div class="lab-top">
<div class="seg" role="group" aria-label="Model">
<button type="button" aria-pressed="true" data-model="a">current: model-a</button>
<button type="button" aria-pressed="false" data-model="b">candidate: model-b</button>
</div>
<div class="score mono">
<span>Pass rate<b id="rate">96.8%</b></span>
<span>Regressions<b id="regs">0</b></span>
<span>Cost / 1k runs<b id="cost">$4.10</b></span>
</div>
</div>
<table class="tbl">
<thead><tr><th>Test case</th><th class="hide">Customer segment</th><th>Grade</th><th>Verdict</th></tr></thead>
<tbody id="rows"></tbody>
</table>
</div>
</div>
</section>
<section>
<div class="wrap">
<h2 class="rv">Fast enough to run on every pull request.</h2>
<div class="counts">
<div class="rv"><b data-count="1.2" data-dec="1">1.2<small>B</small></b><p>tokens evaluated every day across customer workloads</p></div>
<div class="rv"><b data-count="340">340<small>ms</small></b><p>median time to the first verdict on a pull request</p></div>
<div class="rv"><b data-count="97.4" data-dec="1">97.4<small>%</small></b><p>agreement between Orrery graders and expert human reviewers</p></div>
<div class="rv"><b data-count="61">61<small>%</small></b><p>fewer production incidents in the first two quarters</p></div>
</div>
</div>
</section>
<section id="how">
<div class="wrap">
<h2 class="rv">Three lines in CI. Then it watches every change.</h2>
<div class="steps">
<div>
<div class="step rv"><span>01</span><div><h3>Point it at your traces</h3><p>Orrery samples real production conversations into a test set, grouped by customer segment, and keeps it fresh as usage changes.</p></div></div>
<div class="step rv"><span>02</span><div><h3>Write graders in plain language</h3><p>Describe what good looks like. Orrery turns it into a grader, checks it against your own labels, and tells you where it disagrees.</p></div></div>
<div class="step rv"><span>03</span><div><h3>Block the merge on a regression</h3><p>Every pull request that touches a prompt, model, or tool gets a verdict, a diff of the worst cases, and the cost change.</p></div></div>
</div>
<pre class="rv" aria-label="Example CI configuration"><span class="c"># .github/workflows/evals.yml</span>
<span class="k">- name:</span> Run Orrery
<span class="k">uses:</span> <span class="s">orrery/eval-action@v2</span>
<span class="k">with:</span>
suite: <span class="s">support-assistant</span>
compare: <span class="s">main</span>
fail-on: <span class="s">regression > 0.5%</span>
<span class="c"># verdict, 340ms
# 412 cases, 3 regressions (refunds, enterprise)
# cost per 1k runs: $4.10 -> $2.35</span></pre>
</div>
</div>
</section>
<section id="trust">
<div class="wrap">
<h2 class="rv">Your traces stay yours.</h2>
<div class="trust">
<div class="rv"><h3>No training on your data</h3><p>Traces and grades are used to run your evals and nothing else, by contract.</p></div>
<div class="rv"><h3>EU or US residency</h3><p>Pick the region at sign-up. Data never leaves it, including backups.</p></div>
<div class="rv"><h3>PII redacted on ingest</h3><p>Names, emails, and card numbers are masked before a trace is stored.</p></div>
<div class="rv"><h3>SOC 2 Type II</h3><p>Audited every year. The report is available under NDA.</p></div>
</div>
</div>
</section>
<section id="pricing">
<div class="wrap">
<h2 class="rv">Priced on eval runs. Never on seats.</h2>
<div class="plans">
<div class="plan rv"><h3>Free</h3><div class="p">$0</div><ul><li>10,000 eval runs a month</li><li>One project</li><li>Pull request checks</li></ul><a href="#" class="btn"><span data-scramble>Get an API key</span></a></div>
<div class="plan hi rv"><h3>Team</h3><div class="p">$400<small> / month</small></div><ul><li>500,000 eval runs a month</li><li>Unlimited projects</li><li>Trace sampling by segment</li><li>Slack verdicts</li></ul><a href="#" class="btn solid"><span data-scramble>Start a trial</span></a></div>
<div class="plan rv"><h3>Enterprise</h3><div class="p">Custom</div><ul><li>Self-hosted graders</li><li>Single sign-on and audit log</li><li>Dedicated region</li></ul><a href="#" class="btn"><span data-scramble>Talk to us</span></a></div>
</div>
</div>
</section>
<section class="end">
<div class="wrap">
<h2 class="rv">Know before your customers do.</h2>
<div class="ctas"><a href="#" class="btn solid"><span data-scramble>Start free</span></a><a href="#" class="btn"><span data-scramble>Read the docs</span></a></div>
</div>
</section>
<footer>
<div class="wrap">
<span>© 2026 Orrery Labs. A template, the company is fictional.</span>
<nav><a href="#">Docs</a><a href="#">Changelog</a><a href="#">Status</a><a href="#">Security</a><a href="#">Careers</a></nav>
</div>
</footer>
<script>
(function(){
var reduce = matchMedia('(prefers-reduced-motion: reduce)').matches;
document.documentElement.classList.add('js');
/* hero canvas: test cases orbiting, some flip to regressions */
var c = document.getElementById('sky'), x = c.getContext('2d'), W, H, dpr, pts = [], run = true;
function size(){ dpr = Math.min(devicePixelRatio || 1, 2); W = c.clientWidth; H = c.clientHeight; c.width = W * dpr; c.height = H * dpr; x.setTransform(dpr, 0, 0, dpr, 0, 0); }
size(); addEventListener('resize', size);
var rings = [0.16, 0.27, 0.39, 0.52, 0.66];
for (var i = 0; i < 260; i++) {
var r = rings[i % rings.length];
pts.push({ r: r + (Math.random() - .5) * 0.03, a: Math.random() * Math.PI * 2, s: (0.0006 + Math.random() * 0.0012) * (i % 2 ? 1 : -1) / (r * 3), k: Math.random() < .05 ? 2 : Math.random() < .08 ? 1 : 0, z: 0.6 + Math.random() * 1.6 });
}
function frame(t){
var cx = W * 0.68, cy = H * 0.42, R = Math.max(W, H) * 0.62;
x.clearRect(0, 0, W, H);
x.lineWidth = 1;
rings.forEach(function(r, i){ x.strokeStyle = 'rgba(232,236,241,' + (0.05 + i * 0.012) + ')'; x.beginPath(); x.ellipse(cx, cy, r * R, r * R * 0.42, -0.18, 0, Math.PI * 2); x.stroke(); });
x.fillStyle = '#5cf2d6'; x.beginPath(); x.arc(cx, cy, 5, 0, 7); x.fill();
for (var i = 0; i < pts.length; i++) {
var p = pts[i]; if (!reduce) p.a += p.s;
var ca = Math.cos(p.a), sa = Math.sin(p.a), ex = p.r * R * ca, ey = p.r * R * 0.42 * sa;
var px = cx + ex * Math.cos(-0.18) - ey * Math.sin(-0.18), py = cy + ex * Math.sin(-0.18) + ey * Math.cos(-0.18);
var col = p.k === 1 ? '255,107,107' : p.k === 2 ? '155,140,255' : '92,242,214';
var front = 0.45 + 0.55 * (sa + 1) / 2;
x.fillStyle = 'rgba(' + col + ',' + (p.k ? 0.95 : 0.55 * front) + ')';
x.beginPath(); x.arc(px, py, p.z * (p.k ? 1.9 : 1) * front, 0, 7); x.fill();
if (p.k === 1) { x.strokeStyle = 'rgba(255,107,107,.35)'; x.beginPath(); x.arc(px, py, 7 + 3 * Math.sin(t / 300 + i), 0, 7); x.stroke(); }
}
if (!reduce && run) requestAnimationFrame(frame);
}
requestAnimationFrame(frame);
new IntersectionObserver(function(e){ var was = run; run = e[0].isIntersecting; if (run && !was && !reduce) requestAnimationFrame(frame); }).observe(c);
if (!reduce) setInterval(function(){ var p = pts[(Math.random() * pts.length) | 0]; p.k = p.k === 1 ? 0 : Math.random() < .5 ? 1 : p.k; }, 700);
/* crosshair cursor with coordinates */
var H1 = document.getElementById('xhH'), V1 = document.getElementById('xhV'), L = document.getElementById('xhL');
addEventListener('pointermove', function(e){
if (e.pointerType !== 'mouse') return;
document.body.classList.add('xh-on');
H1.style.top = e.clientY + 'px'; V1.style.left = e.clientX + 'px';
L.style.left = (e.clientX + 12) + 'px'; L.style.top = (e.clientY + 10) + 'px';
L.textContent = 'x ' + e.clientX + ' y ' + (e.clientY + scrollY);
});
document.addEventListener('pointerleave', function(){ document.body.classList.remove('xh-on'); });
/* scramble buttons */
var glyphs = '01#%&*+=<>/\\ABCDEFXYZ';
document.querySelectorAll('.btn').forEach(function(b){
var el = b.querySelector('[data-scramble]'), txt = el.textContent, timer;
el.style.minWidth = el.offsetWidth + 'px';
b.addEventListener('mouseenter', function(){
if (reduce) return;
var f = 0; clearInterval(timer);
timer = setInterval(function(){
el.textContent = txt.split('').map(function(ch, i){ return ch === ' ' ? ' ' : i < f / 2 ? ch : glyphs[(Math.random() * glyphs.length) | 0]; }).join('');
if (++f > txt.length * 2) { clearInterval(timer); el.textContent = txt; }
}, 28);
});
});
/* the model swap table */
var cases = [
['Refund over 90 days, annual plan', 'Enterprise', [0.94, 0.41]],
['Cancel and keep data export', 'SMB', [0.91, 0.93]],
['Invoice in a second currency', 'Mid-market', [0.88, 0.9]],
['Escalate a legal threat to a human', 'Enterprise', [0.97, 0.52]],
['Reset two-factor, no phone', 'SMB', [0.86, 0.95]],
['Partial refund after a price change', 'Mid-market', [0.9, 0.44]]
];
var tb = document.getElementById('rows');
function render(m){
var k = m === 'a' ? 0 : 1, regs = 0;
tb.innerHTML = cases.map(function(cs){
var s = cs[2][k], prev = cs[2][0], bad = k === 1 && s < prev - 0.2; if (bad) regs++;
var ok = s >= 0.75;
return '<tr><td>' + cs[0] + '<div class="bar"><i class="' + (ok ? '' : 'lo') + '" style="width:' + Math.round(s * 100) + '%"></i></div></td><td class="hide who">' + cs[1] + '</td><td class="mono">' + s.toFixed(2) + '</td><td><span class="pill ' + (ok ? 'pass">pass' : 'fail">regressed') + '</span></td></tr>';
}).join('');
document.getElementById('rate').textContent = k ? '89.1%' : '96.8%';
var rg = document.getElementById('regs'); rg.textContent = regs; rg.className = regs ? 'bad' : '';
document.getElementById('cost').textContent = k ? '$2.35' : '$4.10';
}
document.querySelectorAll('.seg button').forEach(function(b){
b.addEventListener('click', function(){
document.querySelectorAll('.seg button').forEach(function(o){ o.setAttribute('aria-pressed', o === b); });
render(b.dataset.model);
});
});
render('a');
/* reveals and counters */
function count(el){
var to = parseFloat(el.dataset.count), dec = +(el.dataset.dec || 0), small = el.querySelector('small').outerHTML, t0 = performance.now();
function step(t){ var p = Math.min(1, (t - t0) / 1400), v = to * (1 - Math.pow(1 - p, 3)); el.innerHTML = v.toFixed(dec) + small; if (p < 1) requestAnimationFrame(step); }
if (reduce) el.innerHTML = to.toFixed(dec) + small; else requestAnimationFrame(step);
}
var io = new IntersectionObserver(function(es){ es.forEach(function(e){ if (!e.isIntersecting) return; e.target.classList.add('in'); var b = e.target.querySelector('[data-count]'); if (b) count(b); io.unobserve(e.target); }); }, { threshold: .2 });
document.querySelectorAll('.rv').forEach(function(el){ io.observe(el); });
})();
</script>
</body>
</html>
Once the site is live, connect it to Contengi through a signed webhook. Its agents research, write, and publish articles to the site in your brand’s voice, and keep the old ones fresh.
How the website connection works
A dark studio site where the name fills the screen in condensed capitals made of moving sea water, followed by case studies with parallax photos and hard outcome numbers.
Agencies and studios · 21 November 2026

A two-sided marketplace site for kitchen shift work, with a toggle that rewrites the hero for kitchens or cooks, a photo split into five sliding slices, and shift tickets you could claim tonight.
Platforms and marketplaces · 21 November 2026

An editorial one-person site for a fractional CMO, with a headline that slides in line by line, a manifesto whose words light up as you scroll, and case studies that turn the cursor into a tomato-red read label.
Freelance and fractional · 21 November 2026
FAQ
Yes. Every template is free to use, edit, and ship on personal and commercial projects, with no sign-up and no credit required. Photos come from Pexels under the Pexels licence, so keep the photographer credits for any photo you leave in.
One HTML file with the CSS and JavaScript inside it. There is no build step, no framework, and no package to install: open it in a browser to see it, edit it in any text editor, and upload it to any host to put it live.
Every template comes with a remix prompt. Download the HTML, paste it into Claude, ChatGPT, Cursor, or any coding agent together with the prompt, and fill in the bracketed parts with your name, offer, and copy. The prompt tells the agent what to keep, so the motion and layout survive the rewrite.
Sites for SaaS companies, AI products, freelancers and fractional leaders, platforms and marketplaces, agencies, startups, consultancies, and creators. Each template is built around what that kind of site needs, and each uses its own layout idea, button style, cursor effect, and scroll effect.
Yes. Contengi publishes to sites built with Claude, v0, or any custom code through a signed webhook, so the articles your agents write land on the site you built from the template. The website integrations page explains the setup.
Yes. Each template is checked at six screen widths, from a 375px phone to a 2560px monitor, and cannot be published if anything scrolls sideways or text is cut off. Animations switch off for visitors who ask their device for reduced motion.