Content
Scan old posts for throttle risk before anything is deleted
Old posts you forgot about may still be hurting reach.
What hurt
mr fundman shared a simple use case: have your Grok Bot go through old posts and flag anything that might throttle the account. Related operators described scanning then deleting risky posts. The useful part for a catalog page is the review list — not unsupervised nukes.
What to install or create
Connect X read access. Ask for a FLAGGED vs KEEP list with reasons (policy risk, dead links, off-brand pile-ons). You delete or archive one by one. No bulk delete routine on day one.
Bots
Post risk scanner
FLAGGED vs KEEP with reasons. Never deletes.
Plugins
X
Required
Routines
Optional later: a monthly scan that only refreshes the flag list. Still no deletes.
The prompt
Paste for a scan-only pass
Curated from mr fundman’s published use case in the cited post. Delete stays on the human side.
Last line is our approval fence: Never send, post, spend, or buy without my explicit yes.
Go through my old posts and list anything that might throttle or hurt the account. Return two lists: FLAGGED (with a short reason) and KEEP. Do not delete, hide, mute, edit, or reply. I will act one post at a time. Never send, post, spend, or buy without my explicit yes.
First run
- 1Connect the X plugin with read access.
- 2Paste the scan-only prompt.
- 3Review FLAGGED with reasons.
- 4Delete or keep each post yourself.
- 5Skip bulk-delete routines until you trust the flags.
Approval fence
Do not delete, hide, mute, or edit any post until you confirm each one. No bulk delete.
What to tweak
Steal the scan. Keep the delete button. Same spirit as reply-guard: flags only until you say yes.
Source
mr fundman @mrfundman · Original post
Not affiliated with xAI. We did not write this prompt. Third-party template terms still apply.