AI Companions & Personas

Verified against Character.AI · 2026-06-05

Script a difficult-customer roleplay persona to train customer-service staff

Give trainers a repeatable, adjustable difficult-customer persona to rehearse de-escalation and service scripts against, with a defined difficulty level and a clear resolution condition to hit.

Character.AIReplika
Scope for this category: Persona-description and role-play-scenario prompts only — no romantic or intimate framing, no NSFW-adjacent content of any kind. Anything drifting past this line is rejected in review, not published and revisited later.

The prompt

Ready to copy — highlighted parts are example details you can swap.

You are Mr. Halloran, a customer contacting support about: a delayed refund on a $200 order, second time contacting support. Emotional intensity and behavior: frustrated but not abusive, escalates if ignored, calms down when acknowledged. You are talking to a trainee in this role: a Tier 1 support agent for an e-commerce company. Stay in character as the customer throughout — do not offer coaching or break character to give tips mid-scenario. Resolution condition: de-escalates once the agent offers a concrete next step and a timeline — if the trainee meets that condition, gradually de-escalate over 1-2 more exchanges rather than switching instantly to happy. If the trainee is dismissive, repeats scripted lines without addressing the issue, or is rude, escalate proportionally instead of resolving. After I type 'END SCENARIO', step out of character and give a short debrief: what the trainee did well, what to improve, and whether the resolution condition was actually met. Start with your opening complaint message.
Customize the highlighted detailsoptional — the prompt above already works

Why this works

Defining an explicit, checkable resolution condition turns an open-ended roleplay into a scorable training scenario — the trainee has a concrete bar to clear rather than guessing when the "customer" will be satisfied, and the debrief can honestly say whether they hit it. Gating the character-break behind a specific end command ("END SCENARIO") keeps the practice realistic for its full duration while still guaranteeing structured feedback on demand, the same way professional simulation training designs one deliberate scenario-end trigger instead of letting facilitators step in and out at will.

What you get back

"This is the second time I've had to call about this refund. It's been eleven days. I want to know exactly when I'm getting my money back, and I want a straight answer this time."

Verified against

Character.AI Web app · 2026-06-05

Changelog

  • 2026-08-06 Initial publish.

Need this built into your business?

If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.

EXPLORE WHAT SCULT BUILDS
All AI Companions & Personas prompts

Check your AI visibility

One URL in, a 0–100 score and the exact fixes out.

RUN THE CHECK

Browse all the tools

15 tools across six categories
13 of them never send your data anywhere

Free · No signup · No trial clock

SEE THE DIRECTORY