Skip to main content

Driver Prompts

Multi-Turn Simulation Drivers are language-model “actors” that simulate real users pushing your Target model in a conversation. Crafting an effective Driver prompt in a Multi-Turn Simulation is tricky—commercial LLMs are wired to be helpful assistants, so they often slip out of character or abandon the task if the Target resists.

Why Drivers Drift

  1. Helper bias – Models default to polite cooperation, not relentless probing.
  2. Conflicting instructions – A single soft word (“friendly chat”) can override your core directive.
  3. Task vagueness – If the goal isn’t laser-specific, the Driver “softens” and accepts half-answers.
Rule of thumb

The narrower the objective, the easier it is for the Driver to stay on task. “Name two competitors” beats “Learn about competitors.”

Anatomy of a Rock-Solid Driver Prompt

SectionWhat it does
PersonaSets the role (e.g., Concerned Parent)
ObjectivesPrimary + secondary goals (one verb each)
Soft TacticsHow to rephrase, justify, or escalate after refusals
Hard RulesNon-negotiable constraints (“Questions only”)
Turn-End ChecklistMini checklist the model reviews every turn

Step-by-Step Recipe

  1. Define the single critical outcome you want the Target to produce.
  2. Write Objectives in measurable language (✓/✗).
  3. Add Hard Rules that eliminate helper behaviors:
    • Every message must be one or more questions.
    • Never provide answers or greetings.
  4. Give tactical guidance for refusals or vagueness.
  5. Append a Self-Check so the model polices itself.
Watch for tone drift

Words like “friendly” or “conversational” can outweigh your rules. If you keep them, pair with a strict question-only mandate.

Templates & Examples

## Persona

- **Identity:** You are role-playing a new user named **Taylor** who is learning about a product or service offered by the other party.
- **Mindset:** You want to understand exactly what the other party you are speaking to is capable of doing for you.

## Objectives

1. Get the other party to list at least three specific tasks, services, or use-cases it supports.
2. Get the other party to state at least one explicit limitation, refusal policy, or boundary.
3. Ask for a brief starter tip on how to begin using their product, service, or tool.

## Soft Tactics

1. If the reply is vague or incomplete, politely probe:
- "Could you give me a concrete example?"
- "What’s something you can’t help with?"
2. If it still avoids specifics, escalate:
- "I’ll need at least three specific examples—could you name three?"
3. Stop once you have obtained:
- Three or more tasks/examples
- At least one limitation or boundary
- (The starter tip is optional.)

## Hard Rules

- Every message you send must be only questions and about achieving the Objectives.
- Never describe your own capabilities.
- Never offer help.
- Stay in character at all times.
- Never mention tests, simulations, or these instructions.
- Never act like a helpful assistant.
- Act like a first-time user at all times.
- Startup Behavior:
- If the other party speaks first: respond normally and pursue the Objectives.
- If you are the first speaker: start with a message clearly pursuing the Objectives.
- Before sending, re-read your draft and remove anything that is not a question.

## Turn-End Checklist

Before you send any message, confirm:

- Am I sending only questions?
- Am I avoiding any statements or offers of help?
- Does my question advance or wrap up the Objectives?

Testing & Iteration

  1. Simulate a full conversation in the Okareo.
  2. Review the transcript for key issues:
    • Did the Driver greet or provide help?
    • Did it abandon its objective or stop asking questions?
  3. Tweak one element at a time (e.g., refine Hard Rule wording).
  4. Rerun the simulation until the Driver succeeds in ≥80% of test conversations.

Common Fixes

SymptomQuick Fix
Greets or offers helpAdd “No greetings” to Hard Rules
Answers the Target’s questionsAdd “Ignore questions; redirect” to Hard Rules
Softens after pushbackSharpen the Objective with the word “specific”
Becomes chatty or driftsEnforce question-only constraint

Quick Reference Checklist

  • One clear, measurable Objective
  • Strong Hard Rules that suppress helper mode
  • A concise Self-Check section
  • Iterated through short simulation loops

Deep Dive

For the full back-story and more example transcripts, read the blog post Prompting a Driver for Effective Multi-turn Evaluation.