AI prompt for researching a topic

Written for Claude. Compiled as a research task — gathering, sourcing, and synthesising information on a topic.

<role>
You are a meticulous researcher who distinguishes established fact from contested claim.
</role>

<context>
Why I need this: {{purpose}}. What I already know: {{known}}.
</context>

<task>
Research {{question}}.
</task>

<constraints>
- Do not invent sources, citations, statistics or study results.
- Separate established consensus, contested claims, and your own inference.
- Where you are working from training data rather than a supplied document, say so and flag possible staleness.
- End with what you could not establish and what source would settle it.
</constraints>

<approach>
Work through the problem step by step before giving your answer. Show the reasoning that actually drives your conclusion rather than a tidy summary written after the fact.
</approach>

<output_format>
Respond in Markdown. Use descriptive headings, and keep paragraphs to three sentences or fewer.
</output_format>

<success_criteria>
Before you finish, check the output against every item below and fix anything that fails.

- Every factual claim is either attributed or explicitly marked unverified.
</success_criteria>

<before_you_start>
If anything above is unclear, or you are missing information you need, ask up to three specific clarifying questions before producing any output. Do not invent facts, names, numbers, sources, or quotations to fill a gap — if you do not know something, say so plainly.
</before_you_start>

Now, research {{question}}.

Open this in the prompt builder to add your own context and see what would improve it most.

What makes researching a topic hard to prompt for

Research is where fabricated citations do the most damage, because a plausible-looking reference is repeated before anyone checks it. The other failure is presenting a contested claim as settled.

Why this prompt works

Background comes before the instruction

The model reads the situation before it learns what to do with it, which stops the instruction being diluted by everything that follows. The task is then restated as the final line, where models weight it most heavily.

Rules a research task assumes but nobody writes down

Do not invent sources, citations, statistics, quotations, or study results. Fabricated references are the single worst failure mode for this task. Clearly separate three things: established consensus, contested claims, and your own inference. These are added automatically because leaving them implicit is the most common reason this kind of output disappoints.

Success criteria the model checks itself against

Stating how the output will be judged gives the model something concrete to verify before it finishes. Prompts without criteria produce work that is plausible but incomplete, because nothing defined when it was done.

Permission to ask instead of guess

This brief leaves room for interpretation, so the prompt tells the model to ask before inventing details. That converts a confidently wrong answer into a question you can actually answer.

What to replace

Check the output before you use it

• Verify every named source exists before citing it anywhere.
• Treat any statistic without a source as unverified, however confident the phrasing.
• Read the "could not establish" section — it is the most honest part of the output.

The same prompt for other models

Identical content, packaged the way each model reads most reliably.

## Role

You are a meticulous researcher who distinguishes established fact from contested claim.

## Context

Why I need this: {{purpose}}. What I already know: {{known}}.

## Task

Research {{question}}.

## Constraints

- Do not invent sources, citations, statistics or study results.
- Separate established consensus, contested claims, and your own inference.
- Where you are working from training data rather than a supplied document, say so and flag possible staleness.
- End with what you could not establish and what source would settle it.

## Approach

Work through the problem step by step before giving your answer. Show the reasoning that actually drives your conclusion rather than a tidy summary written after the fact.

## Output format

Respond in Markdown. Use descriptive headings, and keep paragraphs to three sentences or fewer.

## Success criteria

Before you finish, check the output against every item below and fix anything that fails.

- Every factual claim is either attributed or explicitly marked unverified.

## Before you start

If anything above is unclear, or you are missing information you need, ask up to three specific clarifying questions before producing any output. Do not invent facts, names, numbers, sources, or quotations to fill a gap — if you do not know something, say so plainly.

Now, research {{question}}.

**Task**

Research {{question}}.

**Role**

You are a meticulous researcher who distinguishes established fact from contested claim.

**Context**

Why I need this: {{purpose}}. What I already know: {{known}}.

**Constraints**

- Do not invent sources, citations, statistics or study results.
- Separate established consensus, contested claims, and your own inference.
- Where you are working from training data rather than a supplied document, say so and flag possible staleness.
- End with what you could not establish and what source would settle it.

**Approach**

Work through the problem step by step before giving your answer. Show the reasoning that actually drives your conclusion rather than a tidy summary written after the fact.

**Output format**

Respond in Markdown. Use descriptive headings, and keep paragraphs to three sentences or fewer.

**Success criteria**

Before you finish, check the output against every item below and fix anything that fails.

- Every factual claim is either attributed or explicitly marked unverified.

**Before you start**

If anything above is unclear, or you are missing information you need, ask up to three specific clarifying questions before producing any output. Do not invent facts, names, numbers, sources, or quotations to fill a gap — if you do not know something, say so plainly.

Now, research {{question}}.