Read their guide
Made and published by shawn. Start here; it is their original, not ours.
Guide · by shawnHis article with the prompts for all five botsHead of AI Visibility, Founding Content Engineer, GTM Engineer, X API Engineer and Chief of Staff, each with a copy-ready prompt.Open it ↗Example · by shawnHis follow-up: the monthly invoice promptHow the bot found the subscription to replace, and what it costs him now.Open ↗Tools they used · by TregTregThe pay-per-call AI visibility data he used instead of a subscription. He says each tracked prompt cost $0.004.Open ↗
Their prompt
From shawnbuilds Their exact wording: the prompt he runs every month, from his Oct 16 follow-up post. Source ↗
Check my invoices in Gmail. If there are any subscriptions you think you could compete with, design and launch a Grok Bot squad to compete with it based on its 1-2 core features and price. Feed the results to a separate Eval Bot with blinded test to prevent bias. Use any tools/plugins needed, and report back to me if any are fruitful.
What they did
- Runs one prompt monthly: check Gmail invoices, pick subscriptions a bot squad could match on 1–2 core features and price, and test the result with a separate blinded eval bot.
- It flagged Profound ($399/mo), a tool for getting mentioned by ChatGPT and Claude.
- Head of AI Visibility bot pulls AI-engine answer data through Treg and reports which categories you win and lose.
- Founding Content Engineer bot studies why competing articles get cited and writes new ones; Claude Opus writes on Grok Bot's computer and returns a draft he reviews.
- GTM Engineer bot drafts pitches overnight to the directories, blogs and GitHub lists that AI engines cite; he sends them.
- X API Engineer bot gathers daily mentions of his brand and competitors so he can reply himself (5–10 minutes a day).
- A Chief of Staff bot coordinates the four; he talks to it by phone or voice.
- Results he reports: +28% inbound from ChatGPT and Claude after a week; about 80% of Profound's quality; "$8 a month" in a reply and "$45 vs. $399" in his follow-up.
- Heads-up: the setup leans on Treg, which he promotes throughout. We couldn't tell whether he's affiliated. All numbers are self-reported.
“I just killed my Profound subscription ($399/month) with Grok Bot”— shawn (@shawnbuilds), X
Summary written by Agent Cookbook. Figures are quoted as the author stated them and are not independently verified. See the original for full context.
Approval rule
🛡
Articles stay drafts until you review them, and outreach pitches wait for you to send. Sending, buying, posting, and deleting always need a human.
Set it up
- Run his monthly prompt in your main Grok Bot and pick one subscription it flags; don't try to replace several at once.
- For an AI-visibility tool like his, open his article and create the four specialist bots from his prompts, then the Chief of Staff bot that talks to them.
- Connect a pay-per-call data source (he used Treg) and have the visibility bot report wins and gaps by category.
- Have the content bot draft 2–4 articles a week for the gaps (his pace); you review each one before it goes live.
- Let the GTM bot draft outreach to the pages AI engines cite; you read and send every pitch yourself.
- Keep the old subscription until the blinded eval says the bots are close enough, then cancel.
Common failures
⚠
- Expecting day-one parity: he says the first versions were about 10% as good and needed a lot of feedback.
- Letting the outreach bot send on its own. Keep pitches as drafts.
- Writing one article a month and expecting citations to move; he says consistency wins.
Variations
- Point the monthly invoice prompt at other tools: social schedulers, report builders, monitoring.
- Swap Treg for any API that returns AI-engine answers per prompt.
Get the next money recipe. Free.
New recipes every week, plus the top 10 agent setups pack.
Unsubscribe anytime. Privacy