コンテンツ

Flag burner replies and insults

A big account's replies are a firehose of burners and slurs.

困っていたこと

Jason Calacanis built a Grok Bot that reviews replies on his posts. It flags suspected burner accounts (under 100 followers, or under a year old) and replies that use insults. He said he wants the thread classy, fun, and intelligent, and that he would auto-nuke under-one-year-old accounts. We keep hide, mute, block, and reply behind your yes. The published prompt already does that.

インストールまたは作成するもの

One screening bot. Connect the X plugin so it can read replies. Paste the published prompt. Point it at a post. It returns two lists: FLAGGED (burner / new account / insult) and CLEAN (including people who just disagree). Insults are contempt aimed at a person, not heat. When it is unsure, it keeps the reply in CLEAN and tells you why.

ボット

Reply Guard
Two lists. Flags only. You act.

プラグイン

X
必須

ルーティン

Optional, after you trust the flags: a check on new replies to your latest post. Still no hide, mute, block, or reply without you.

プロンプト

Paste into Edit Profile

Published prompt reconstructed from @Jason's public setup. Not his verbatim text.

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.

初回実行

1Create a Grok Bot. Name it Reply Guard or whatever you will actually open.
2Paste the prompt into Edit Profile. Edit the follower floor and account-age cutoff if Jason's numbers are not yours.
3Connect the X plugin so it can read replies on your account.
4First task: paste one of your own post URLs. Read both lists. Spot-check two FLAGGED and two CLEAN rows.
5Only after the calls look right, consider a routine on new replies. Keep hide, mute, block, and reply on your side.

承認フェンス

The published prompt already says the action is yours. Jason talked about auto-nuking new accounts. On this page, nothing is hidden, muted, blocked, or answered until you say yes.

調整ポイント

Copy the split, then set your own floors. Jason's 100 followers and one-year cutoff are his. Change them in the prompt before you trust a flag. Disagreement stays clean on purpose. Do not let the bot hide, mute, block, or reply, even if you later decide to automate nukes. All bots share one computer. A reply-guard that can also post is a different, riskier setup. The content studio on /teams is the publishing roster with the same gate.