# DenQAI: ai prompt pack

File-specific companion · 2026-09-10

## Purpose

How can AI help with a bounded dental-practice decision without inventing facts or assuming professional authority?

**Who:** Owner and accountable subject-matter reviewer

**Records:** Permitted coded or aggregate records; current primary-source excerpts; defined question; source locators; model assumptions and units.

## Complete the file

1. Apply Evidence guardrail before selecting the relevant task prompt.
2. Provide only authorized inputs and use the prompt for market census, transfer, fees, clinical-review planning, billing-review planning, role evidence, or red-team analysis.
3. Require citations, separated facts/assumptions/calculations, missing-evidence questions, and decision-reversal conditions.
4. Have the appropriate human reviewer check the output before using it in a decision.

## Definitions

- **Direct record:** Source material documenting the particular event or amount, still subject to verification.
- **Inference:** Reasoned interpretation that extends beyond what the source states directly.
- **Decision-reversal condition:** Fact or assumption change that would alter the proposed decision.

## Fictional example

Fictional example: Give AI a fictional five-line collections bridge and ask it to identify overlap and cite the supplied row for every adjustment; missing amounts remain questions.

## Review and next action

Use the AI output to request missing evidence or test a calculation, then record the human-reviewed conclusion.

This prompt pack is a readable aid, not a secure AI environment or professional review. Do not upload PHI or restricted deal data to an unapproved service or let AI invent work, diagnoses, codes, or contract terms.

[Related DenQAI guide](https://denqai.com/ai-workflows)

## AI coach prompt

Help me use DenQAI’s ai prompt pack. My question is: How can AI help with a bounded dental-practice decision without inventing facts or assuming professional authority? Explain these terms in plain language: Direct record, Inference, Decision-reversal condition. Ask one question at a time about the required records: Permitted coded or aggregate records; current primary-source excerpts; defined question; source locators; model assumptions and units. Walk through the supplied fictional example before asking me for inputs. Treat every missing value as unknown, not zero. Do not invent source records, contract terms, legal conclusions, or clinical facts. Use only permitted fictional or aggregate, non-identifying information. Follow this boundary: This prompt pack is a readable aid, not a secure AI environment or professional review. Do not upload PHI or restricted deal data to an unapproved service or let AI invent work, diagnoses, codes, or contract terms. End with the missing evidence and this next step: Use the AI output to request missing evidence or test a calculation, then record the human-reviewed conclusion.

## AI review prompt

Review my permitted fictional or aggregate draft of DenQAI’s ai prompt pack against the file instructions. Check each required field and period, units, evidence locator, assumptions, arithmetic where formulas are actually supplied, and unresolved contradictions. Check the sequence: Apply Evidence guardrail before selecting the relevant task prompt. Provide only authorized inputs and use the prompt for market census, transfer, fees, clinical-review planning, billing-review planning, role evidence, or red-team analysis. Require citations, separated facts/assumptions/calculations, missing-evidence questions, and decision-reversal conditions. Have the appropriate human reviewer check the output before using it in a decision. Separate facts from assumptions and missing records. Do not supply unsupported numbers or make a professional decision for me. Apply this limitation: This prompt pack is a readable aid, not a secure AI environment or professional review. Do not upload PHI or restricted deal data to an unapproved service or let AI invent work, diagnoses, codes, or contract terms. Return corrections, questions for the responsible reviewer, and the next action.