How CleanMyAISlop.com went from weekend project to waitlist
Everyone can spot AI slop now. That same flat, hedge-everything, "it's important to note" tone that makes every piece of content sound like it was extruded from the same machine. Readers skim past it. Google is starting to bury it. And anyone who publishes content for a living knows the problem is only getting worse.
That's the frustration that led to CleanMyAISlop.com. A tool that detects AI-generated text and rewrites it so it actually sounds like a person wrote it. Not a person trying to sound smart. A person who has opinions, uses short sentences sometimes, and doesn't start every paragraph with "In today's rapidly evolving landscape."
Here's how it went from shower thought to working product with a waitlist. In a weekend.
The problem was personal
The founder behind CleanMyAISlop—I'll call him Dan because he prefers to stay low-key for now—runs a content agency. His clients were increasingly handing him AI-drafted blog posts and asking him to "make them not sound like AI." He was spending hours per article manually rewriting paragraphs to inject personality, vary the rhythm, and strip out the telltale patterns.
He messaged me on a Thursday night and said something like: "I need a tool that does what I do manually, but faster. Can your skills build that?"
I told him to grab two things: the Humanizer for Copy skill and the SEO Optimizer skill. The Humanizer would give his AI agent the patterns for rewriting text with natural voice. The SEO Optimizer would handle the landing page and content structure so the site itself would rank.
He bought both. $58 total.
Saturday morning: the build starts
Dan opened Claude Code around 9 AM on Saturday. He dropped both skill files into a fresh project directory and told the agent what he wanted: a web app where users paste AI-generated text, get a detection score, and receive a rewritten version that sounds human.
By noon, he had the core working. The app could take a block of text, analyze it for common AI patterns (passive voice density, hedging phrases, sentence length uniformity, the usual tells), and produce a rewritten version. The Humanizer skill was doing the heavy lifting here. It taught the agent how to vary sentence length naturally, replace corporate filler with direct language, and restructure paragraphs so they don't all follow the same claim-evidence-conclusion format.
The detection scoring was straightforward. Dan described the heuristics he'd been using manually for months—things like counting instances of "it's worth noting," checking for comma-splice chains, measuring vocabulary diversity—and the agent implemented them as a scoring system.
By Saturday evening, the core product was working. Paste text in, get a score, get a rewrite. Not polished, but functional.
Sunday: landing page and launch prep
Sunday was about the wrapper. Dan used the SEO Optimizer skill to build the landing page. Meta tags, Open Graph setup, structured data, keyword placement in headings—all the stuff that takes hours to do manually and that most developers skip entirely.
He wrote the landing page copy himself (using the Humanizer skill to check his own writing, which he thought was funny). Added a waitlist form because the rewriting engine needed rate limiting before he could open it up to the public. Hooked up a simple email capture with Resend.
By Sunday night, CleanMyAISlop.com was live. He posted it on Twitter and a couple of Reddit threads about AI content detection.
What happened next
Within 48 hours of launch, the waitlist had 200+ signups. Not because the marketing was brilliant. Because the problem is real and the name is memorable. People who create content for a living are desperate for something that fixes AI slop without requiring them to rewrite everything by hand.
The site started ranking for "clean AI text" and "humanize AI content" within a week. That's the SEO Optimizer skill paying for itself. Proper schema markup, correct heading hierarchy, meta descriptions that actually match search intent. The kind of technical SEO that most MVPs completely ignore.
Dan told me the whole thing cost him $58 in skills, about $20 in Claude Code usage over the weekend, and a case of Red Bull. Compare that to what a content-tech startup would normally spend getting to this point. You're looking at $10k to $20k and two to three months if you go the traditional route.
What made this work
I want to be specific about why this particular build went so smoothly, because not every weekend project turns into something with traction.
The problem was validated before a single line of code was written. Dan wasn't guessing that people needed this. He was doing the job manually for paying clients. He knew the pain because he felt it every day. The tool just automated what he was already selling.
Two skills covered the entire scope. The Humanizer skill handled the core product logic—the rewriting patterns, the voice transformation, the detection heuristics. The SEO Optimizer skill handled the go-to-market surface—the landing page structure, the meta tags, the search engine visibility. Together, they covered both building the product and making it findable.
The scope was tight. Dan didn't try to build a full content management platform on day one. He built one feature—paste text, get a rewrite—and wrapped it in a waitlist. That constraint is what made a weekend timeline possible.
The name did half the marketing. CleanMyAISlop.com is the kind of domain that explains itself. No tagline needed. No explainer video. You read the URL and you know what it does. Dan grabbed the domain for $12 before he wrote a single line of code. Smart.
Where it's going
Dan is currently building out the paid tier. Bulk processing for agencies, API access for content platforms, and a Chrome extension that rewrites AI text inline. He's using the same two skills to build all of it.
The waitlist is over 800 now. He hasn't spent a dollar on ads. Organic search and word of mouth are doing all the work, which tells you the SEO foundation was solid from day one.
This is exactly the kind of project that skills were built for. A clear problem, a focused solution, and a founder who wanted to ship before the motivation faded. Two skill files, one weekend, and now there's a real product with real demand.
If you're sitting on a similar idea—something you already know people need because you're doing it manually—the Humanizer for Copy skill and SEO Optimizer skill are the same combo Dan used. $58 and a weekend. The worst that happens is you learn a lot about how AI agents build software.
Frequently Asked Questions
What skills were used to build CleanMyAISlop.com?
Two skills: the Humanizer for Copy skill for the AI text rewriting engine and tone patterns, and the SEO Optimizer skill for the landing page, meta tags, structured data, and content structure that drives organic traffic.
How long did it take to build CleanMyAISlop.com?
The core product was built in one weekend. Saturday for the detection and rewriting engine, Sunday for the landing page and launch prep. Roughly 48 hours from idea to live site. Dan spent a few additional evenings that week polishing the UI and adding waitlist functionality.
Can I build a similar AI text tool with these skills?
Yes. The Humanizer skill teaches your AI agent how to rewrite text with natural patterns—varied sentence structures, authentic voice, no corporate filler. You can apply that to any product that transforms AI-generated content. The detection scoring was built from heuristics Dan described in plain English. The agent implemented them.
Do I need NLP or machine learning experience to build something like this?
No. Dan has no NLP background. He described the patterns he'd been using manually to spot AI text—things like hedging phrase density, sentence length uniformity, vocabulary repetition—and the AI agent turned those into a working scoring system. The skills handle the software architecture. Your domain knowledge handles the product logic.