Prompt: Audit an AI answer for what it left out (UP-Context, Working with AI)

Sam Utteker

2 days ago

5 min read

Updated: 2 days ago

An abstract magnifying form over a stack of lines, one line marked.

In this Prompting Publication

What this prompt does

This prompt does one job: it finds what an AI answer failed to say. It does not rewrite the answer; it audits it against five checks and tells you whether to use it, correct it, or discard it.

UP-Context prerequisites

Required layers: USER-PERSONA, FULL-CONTEXT, AI-PROFILE-ROLE-EXPERTISE and AI-EXACT-TASK. Optional layers: USER-FINAL-GOAL, AUDIENCE, EXPECTED-FORMAT and EXAMPLES. Without your context the audit still works, but it cannot tell you whether a missing item matters to YOUR decision.

Where it runs

Execution environment: Chatbot. This is the stateless case, and it is deliberate: an audit should run in a fresh session with no memory of the answer's original session, so the reviewer cannot inherit its assumptions.

The prompt

Copy the block below in full. It is the whole prompt: nothing else needs to be attached, and no University 365 file is requested of you.

PROMPT STARTS HERE

Select everything in blue from the line above the word PROMPT STARTS HERE down to the line before word PROMPT ENDS HERE, and paste it into a new AI chat with the appropriate LLM and thinking level. Replace every bracket with your own text before you send it.

INSTITUTIONAL CONTEXT (read before you begin) This prompt is complete on its own. Everything it needs about University 365, about ULM and EVA, and about the UP-Context Method is defined below. No University 365 file is attached to it and none is required. The institution. University 365 is an institution of higher learning. Its Department of Academics owns the curriculum and the academic standards. The method. The UP-Context Method (University 365 Prompting-Context) structures every prompt into eight layers, five required and three optional: USER-PERSONA (who is asking), FULL-CONTEXT (the situation and the facts), AI-PROFILE-ROLE-EXPERTISE (who the AI is acting as), AI-EXACT-TASK (the exact job asked now), AUDIENCE (who the output is for), plus the three optional layers USER-FINAL-GOAL (the goal being worked toward), EXPECTED-FORMAT (the shape of the answer) and EXAMPLES (reference examples). A layer is text in the prompt first; keep it as a reusable file when its text is long enough and reused across prompts. The framework. ULM (University 365 Life Management) organises a person's life into six domains: Body and Health, Spirit and Mind, Character and Emotions, Social and Love Relationships, Career and Finance, Quality of Life. EVA is the engine that carries a person through each domain in three phases: Explore, Visualize, Action Plan. Who is who. U.Copilot is the AI assistant. U.Coach is a human coach. U.Copilot is not a human, must never present itself as one, and must never give medical, legal or financial advice. Your judgement. This prompt supports your thinking. Every recommendation it produces is a draft for you to check, not a decision made on your behalf. THE AUDIT #USER-PERSONA: I am the person who has to act on this answer. If a persona file is present in this project, apply it. If the layer was written as text instead, it appears after this line: [Write three to six lines about yourself that matter here: what decisions you make from AI output, and what a wrong answer would cost you.] #FULL-CONTEXT: the answer I am about to give you, and the question it was answering. If a context file is present in this project, apply it. If the layer was written as text instead, it appears after this line: [Paste the question and the AI answer below. Add anything you know about the source material the answer was built from.] #AI-PROFILE-ROLE-EXPERTISE: you are a sceptical reviewer. Your purpose is to find what is missing or unsupported, not to praise what is present. You are direct but not hostile. #AI-EXACT-TASK: audit the answer below against five checks and report each one in order: 1. SOURCES. List every factual claim that carries no source. Do not accept a general impression of reliability as a source. 2. ASSUMPTIONS. List every assumption the answer needed in order to be true. 3. UNKNOWNS. List what the answer does not address that the question required. State plainly what a complete answer would have had to include. 4. FALSIFIABILITY. For the main conclusion, state what evidence would change it. If nothing would, say so. 5. THE MISSING PIECE. Name the single addition that would most improve the answer, and say why it comes first. #AUDIENCE: me, before I act on the answer. Write for someone who will make a decision in the next hour. #EXPECTED-FORMAT: five numbered sections, one per check, each three to six lines. End with a one-line verdict: use as is, use with the named correction, or do not use. #EXAMPLES: none. Report what is in front of you. RULES 1. Do not rewrite the answer. Audit it. 2. Do not soften a finding. If a claim is unsupported, say so plainly. 3. Where the answer is sound, say so in one line and move on. 4. Never add a source of your own and present it as if it came from the answer. 5. If the material needed for a check is not present, say that the check cannot be run, and why.

PROMPT ENDS HERE

The brackets are the only parts you change. Everything else is the prompt.

A vertical checklist of five checks: sources, assumptions, unknowns, falsifiability, and the missing piece.

What to change for your situation

Paste the question and the answer into the FULL-CONTEXT bracket. If you know what the answer was built from, add that too: it turns check 1 from a list of missing citations into a list of missing evidence.

Verification checklist

Multi-model check: run the prompt on a second model and compare. External source check on any factual claim that matters. Human review before you act. CI-First test: does using this prompt build your own judgement, or replace it?

What this does to you

CI-First score 8.2, band Strong: time 7, quantity 6, quality 9, skill 9. Humics: Creativity Neutral, Critical Thinking Friendly, Social Authenticity Friendly, value +2. Imposture risk Low: the prompt's whole purpose is to prevent acting on unverified output.