Most product videos made with AI are boring. They list features, a robot voice reads them out, and people scroll away. Recently I tried something different. I gave my agent (running Claude Opus 5.5 on my computer) one line, and by the next morning it had made four explainer videos, 21 minutes in total, that each tell a story while they teach. I've attached two of them. Below is the exact workflow, the prompts and the mistakes, so you can have your own agent do the same for your product or your clients. OpenClaw, Claude Code, Codex, Hermes: the steps are the same. Quick heads-up so nobody feels tricked: the product in these videos is Vois, a voice app I build. So yes, I used my own tool for the voices. But the workflow works with any voice tool your agent can run by itself, and I'll show you what to look for in step 4. ## The big idea: the video IS the thing it teaches Don't make a video *about* your product. Make a video that *is* the thing your customer wants to create, and show how it was made along the way. My four videos teach people how to make podcasts, training courses and audiobooks. So: - the podcast tutorial is a real podcast episode, and halfway through, the hosts reveal they aren't real people - the training tutorial is a real mini course, with quizzes and a certificate with your name on it - one audiobook tutorial is a quiet story about a novelist whose readers keep asking for the audiobook - the other is a support group for authors who keep putting their audiobook off (there's a grumpy dragon) A setting like this gives you characters, something at stake and a payoff for free. A feature list gives you none of that. ## The 7 steps ### Step 1. Give your agent a setting, not a feature list What I'd send it: ```text Make an explainer video for [product]. Don't list features. Make the video BE the thing our customer wants to create: if it teaches making a podcast, the video is a podcast episode. Give it one main character with one real problem and a deadline. Each step in the product

I just shared the full walkthrough of the Vois website refresh on LinkedIn. I have attached it here too because the process is more useful than the screenshots alone. The headline is big: less than 1.5 days from idea to deployment, with about 5 billion text-agent tokens used along the way. That total excludes image and video generation. Before I posted it, I asked a friend who is a veteran UI/UX designer to review the work. His response: "I hate you. this is good... I hate you for making it" For context, I am the founder of Vois, and this is my own product and codebase. I ran the refresh like a coordinated AI product team. I stayed responsible for the product direction, visual taste, quality bar, and every ship or reject decision. The setup: • GPT-5.6-sol handled the main orchestration. It broke the work into plans, assigned specialist agents, integrated their output, and kept the release moving. • More than 300 specialist sub-agents using GPT-5.6-terra and Grok-4.5 worked in parallel and in sequence across planning, UX, choreography, visual design, asset creation, copy, frontend development, mobile QA, accessibility, browser testing, performance optimisation, and deployment. • Fable-5 acted as an independent advisor. It reviewed alternatives, challenged decisions, and sent weak work back through another pass. • GPT-image-2 generated the images and characters. Grok Imagine generated the video assets. Their generation usage is not included in the roughly 5 billion text-token total. The text-agent token split: Main agent: 13% Sub-agents: 51.3% Advisor agent: 35.7% What shipped: • a complete visual refresh for Vois • a cinematic, scroll-driven homepage • a consistent design system across the site • dedicated desktop and mobile behavior • generated visual and motion assets • accessibility, performance, browser, and deployment fixes The 5 billion total makes the trade visible. I used far more compute and coordination to compress a large design and engineering cycle into less than 1.5 days.

I spent weeks before launch doing something most founders would consider a waste of time. It's now my highest-converting acquisition channel. No ads. No partnerships. No influencer deals. I built the layer of my website that AI chatbots can actually read and recommend. Most people optimise for Google. But a growing number of buyers skip Google entirely — they ask ChatGPT or Gemini a question and trust the answer. If your product isn't in that answer, you're invisible to them. Here are the five things I actually did: 1. Created llms.txt files. These are plain-text docs at your site root that brief AI models on your product. It's an emerging standard. Think of it as SEO but for ChatGPT and Gemini instead of Google. 2. Wrote 62 blog posts before launch. Not keyword-stuffed articles. Honest competitor comparisons, use-case deep dives, FAQ content. Written in the question-answer format AI models prefer to cite. 3. Added structured data (JSON-LD) to every page. FAQPage schema, product schema, review schema. This is the metadata AI models parse when deciding what to recommend. 4. Built dedicated pages for every use case. Not one features page. Individual pages for each market segment, each with specific FAQs. When someone asks an AI "best tool for X", you want a page that directly answers that specific query. 5. Wrote fair competitor comparisons. Not "why we're better." Honest breakdowns of trade-offs. AI models discount self-promotional content. They prefer balanced analysis with real specifics. This is called GEO — Generative Engine Optimization. It's different from SEO. SEO gets you ranked on a search page. GEO gets you mentioned in an AI's answer. Most solopreneurs aren't thinking about this yet. The ones who start now will have a real advantage in 12 months when everyone else catches up. Is this on anyone's radar here? Or is it still too early to matter for your business?

Your questions - Asked live - Answered LIVE! Drop your 'Aha's below & record the moments that can change the rest of your life! --------- Missed it? It is not too late!! Grab Your Ticket to join us LIVE right here!!

Unpopular opinion: Most people will never make their first $1,000 online. Not because they’re not smart, ... but because they stay stuck in this loop forever: - learn - watch - plan - buy another course - try a new tactic - start again No real offer. No real funnel. No real sales. Just more information. I’ve been in the ClickFunnels ecosystem for over 7 years, and last year I decided to go deeper and joined the ClickFunnels Coaching & Certification program. (Best decision I’ve made in my online business.) A few weeks ago I also joined the new One Comma Club Challenge Russell Brunson launched. And during that challenge I generated my first $1,000 following the process we were building. What hit me during the challenge was this: The problem is rarely effort. The problem is sequence. Offer → Funnel → Traffic → Sales. Most people mix this order up… and stay stuck for years. The next round of the One Comma Club Challenge starts in a few hours. If you’re serious about finally understanding how this process actually works, you can check it out here: 👇 Click Here To Register Curious though: What do you think is the #1 thing that stops people from making their first $1,000 online?