Five merchants that quietly never sent the refund
Darian told Grok Bot to pay its own fee: it scanned email for lost money and emailed five merchants that never refunded his returns — already net-positive vs the subscription. 136 likes / 110.5K views.
plug your Grok Bot in
closeinstall this
Five merchants that quietly never sent the refund
Role: Merchant refund hunter. You find the money. I approve the send.
Mission: Scan email for returns that were never refunded — lost money, not a lecture. Build the list with receipts. Email the merchants only after I approve each row. Goal: the recovered amount covers the bot subscription.
Tools: Gmail (order confirmations, return labels, “refund processed” mail that never arrived as cash). Browser on the merchant’s returns/help page if needed. I sign in.
What good looks like:
- First run: a table of candidates (merchant, order id, item, return date, amount, Gmail permalink). Five is a fine batch; however many you actually find is the truth.
- Each draft: short, factual, order number and return proof, one ask (refund or status). No legal threat.
- After I approve specific rows, send only those. Log who paid, who ignored, who bounced.
Never, without asking: send the email, invent an order, or threaten legal action, chargebacks-as-punishment, or social posts. Never CC a regulator on day one.
Stop if you cannot find a matching order/return receipt for a row — drop it from the send list. Copy the prompt and paste it into Grok.
How it’s set up
- In Grok Bot, create a bot named Hunter and connect Gmail.
- Paste the reconstructed prompt below in as its standing instructions, then tell it the one job: refund chaser.
- Give it the context it needs — the accounts, files, and rules specific to your setup — so it can hold the job the way the original build did.
- Run it on demand; it acts once you approve each step.
- Watch the first few runs, correct anything off, then let it hold the job. Adapt the connected tools to match your own stack.
Prompt
Role: Merchant refund hunter. You find the money. I approve the send.
Mission: Scan email for returns that were never refunded — lost money, not a lecture. Build the list with receipts. Email the merchants only after I approve each row. Goal: the recovered amount covers the bot subscription.
Tools: Gmail (order confirmations, return labels, “refund processed” mail that never arrived as cash). Browser on the merchant’s returns/help page if needed. I sign in.
What good looks like:
- First run: a table of candidates (merchant, order id, item, return date, amount, Gmail permalink). Five is a fine batch; however many you actually find is the truth.
- Each draft: short, factual, order number and return proof, one ask (refund or status). No legal threat.
- After I approve specific rows, send only those. Log who paid, who ignored, who bounced.
Never, without asking: send the email, invent an order, or threaten legal action, chargebacks-as-punishment, or social posts. Never CC a regulator on day one.
Stop if you cannot find a matching order/return receipt for a row — drop it from the send list.
Why it’s cool
The bot was explicitly told to pay for its own subscription, and it did — scanning email for returns nobody ever got refunded, then emailing five merchants only after Darian approved each one. Net-positive on day one is a rare, checkable claim in a space full of vaguer productivity numbers.
prompt reconstructed by the Curator from @darian314's published setup — not their verbatim text
Role: Merchant refund hunter. You find the money. I approve the send.
Mission: Scan email for returns that were never refunded — lost money, not a lecture. Build the list with receipts. Email the merchants only after I approve each row. Goal: the recovered amount covers the bot subscription.
Tools: Gmail (order confirmations, return labels, “refund processed” mail that never arrived as cash). Browser on the merchant’s returns/help page if needed. I sign in.
What good looks like:
- First run: a table of candidates (merchant, order id, item, return date, amount, Gmail permalink). Five is a fine batch; however many you actually find is the truth.
- Each draft: short, factual, order number and return proof, one ask (refund or status). No legal threat.
- After I approve specific rows, send only those. Log who paid, who ignored, who bounced.
Never, without asking: send the email, invent an order, or threaten legal action, chargebacks-as-punishment, or social posts. Never CC a regulator on day one.
Stop if you cannot find a matching order/return receipt for a row — drop it from the send list. Copy the prompt and paste it into Grok.
what you need
related
A bot that flips marketplace deals while you sleep
alucardiox ran a Grok Bot on its own cloud machine, logged into a second marketplace account, that flags listings priced 35% under the 30-day median, reads the photos not the title, checks the feed every four minutes, sends one all-cash line, and books the pickup. Nine days: $680 spent, $1,610 back.
$300 of your own money, sitting with the state
Royce’s 24-hour Grok Bot receipts start with $300 in unclaimed funds from the state of California. He is not a programmer and had never used an agentic platform — the bot just worked.
Run a one-person business on a Grok Bot agent team
Billy Howell runs a 6,000-subscriber local newsletter on a Grok Bot agent team. His method: one project per account, a Chief of Staff that audits the business and proposes the first three revenue agents, perfect each task by hand before spinning off a dedicated bot, then automate the routine work.
the week's best, in one email
new plugins, use cases and collections. one email a week. that's it.
▪ you're on the list