The $2k-a-month marketing bottleneck a bot took over
Jon ONeill (HouseHackerJon) hired a Grok Bot named Mary as marketing manager for his Phoenix plumbing company.
Marketing·Advanced·0 saves
Setup USE CASEMarketingModerate
“Screen your X replies for burners and insults”
Reviews the replies to your posts and splits the signal from the noise: it flags accounts under 100 followers or under a year old, plus anyone slinging insults, so real conversation stays visible and throwaway replies get moderated. Jason Calacanis built it to keep his mentions classy.
How it runs
4 steps. Do them in order.
- 01Connect your X account (the X plugin) so the bot can read the replies to your posts.
- 02Set your thresholds: the follower floor (Jason uses 100), the account-age cutoff (under a year), and what counts as an insult versus a strong opinion.
- 03Paste the prompt below. Point it at a post; it scores each replier against your rules and returns two lists — the burners and insults separated from the genuine conversation.
- 04Review the flags. Keep hiding, muting or blocking behind your own approval until the calls have earned your trust.
What you get
Most reply moderation is all-or-nothing: you either read everything or you tune it out. This splits the difference. It never judges the *conversation* for you — real disagreement stays in the CLEAN pile on purpose — it just pulls out the two things that are almost never worth your attention: throwaway burner accounts and contempt aimed at a person. The restraint is the whole design: it flags, you decide, and nobody gets auto-nuked on a bot's hunch.
Prompt
Copy it. Change the names. Keep the job.
Prompt
You screen the replies to my X posts. For each reply, look up the replier's account and flag it if it meets any of my rules: fewer than 100 followers, an account created less than a year ago, or a reply that carries a personal insult or slur aimed at me or at another user. Return two lists. - FLAGGED: the reply, a link to it, and the reason (burner / new account / insult). - CLEAN: genuine replies worth engaging, including people who simply disagree with me. Judge insults by intent, not by heat. Disagreement, criticism and strong opinions are CLEAN. Contempt aimed at a person — name-calling, slurs, dehumanising language — is FLAGGED. When you are unsure, put it in CLEAN and tell me why you hesitated. Do not hide, mute, block or reply to anyone on your own. Your job is the flagged list; the action is mine.
Use withGrok
via @jason · Tweet by @Jason
Tips
- ·Point it at your own X account and set your thresholds — the follower floor, the account-age cutoff, and what counts as an insult. It flags for review; keep hiding, blocking or muting behind your approval until you trust the calls.
- ·adhoc
- ·proposes
Related setups
Same category. Different job.
A draft in my voice that flags its own weak claims
Farzad’s Grok Bot team list is mostly names; the job to copy is Writey — article/newsletter writer. He spun the team up to see how it goes; this is the one with a real content job, not “Idea” or “Master”.
Marketing·Moderate·0 saves
Setup Automate your affiliate program with an AI affiliate manager
An Affonso-powered Grok Bot automates weekly affiliate management: it compares performance, finds growth opportunities and risks, and prepares partner actions. Program changes, payouts, and partner messages still require approval.
Marketing·Moderate·0 saves
Setup A voice note becomes a live LinkedIn post
Debbie created a LinkedIn bot, signed in on the bot’s computer (it never sees the password), let it learn her existing posts, dumped a messy voice note, and told it to post. It posted. Receipt, not a draft-only demo.
Marketing·Moderate·0 saves
Setup