The four heights
The same task, four distances: today's deadline, the next reviewer, the stuck moment, the pattern.
Execute — do the immediate task
+Block and ban the user who has been harassing staff: remove them from the Page, ban them from…
Execute — do the immediate task
+Block and ban the user who has been harassing staff: remove them from the Page, ban them from posting or messaging, and log the date and reason. Before banning, archive their recent harassing comments and direct messages, then confirm the ban is active and send me the screenshot of the user’s profile and the ban confirmation.
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Improve — make it easier to accept
+Before I ban this repeat harasser from the Page, make the action hard to contest for reviewers and…
Improve — make it easier to accept
+Before I ban this repeat harasser from the Page, make the action hard to contest for reviewers and easy for my team: show the sequence of harassment with timestamps, include links to repeated violations, note prior warnings we sent, and surface any comments that could be misconstrued as regular feedback. Highlight anything that would make a moderator hesitate and how to address it.
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Decide — diagnose the stuck moment
+A customer repeatedly insulted and threatened staff in comments; I temporarily banned them but the…
Decide — diagnose the stuck moment
+I just banned a user for harassing staff but the community manager is nervous about backlash.
A customer repeatedly insulted and threatened staff in comments; I temporarily banned them but the community manager worries this will escalate or that the user will claim we censored legitimate criticism. I don’t know how strict the platform is about bans or whether we should document more. What is the likely outcome of a permanent ban, what documentation will protect us, and what’s the best next communication to the team and to the complainant if they escalate elsewhere?
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Become — change the pattern
+Over the past six months we issued inconsistent warnings and bans for similar harassment, causing…
Become — change the pattern
+We keep toggling bans and warnings without a clear rule set, wasting time and inviting disputes.
Over the past six months we issued inconsistent warnings and bans for similar harassment, causing staff confusion and occasional public rows. Where are we losing credibility and time, and what policy or short rubric should we adopt so community managers make consistent ban decisions in under five minutes while collecting the right evidence for appeals?
Pasted it? When the reply comes back, push once: ask it to sharpen the weakest part. — Did this prompt help?
Next to this one
Other social platform work people do in Facebook.
Every task here came from the work, not from a feature list — which is why the prompts name what you want done and never the button that does it. The tool changes; the work does not.
Copyright © LLOS.ai · 2026 — original pedagogy, voice, and design — all rights reserved.
The rest of the map
Same library, five ways in.