home/roleplay/research-brief-review

Research Brief Review

GPTClaudeGemini··1,217 copies·updated 2026-07-14
research-brief-review.prompt
# Prompt: Research Brief Review

Use this prompt when you want an AI tool to review a short research brief or memo for unsupported claims, overstatements, and missing caveats. The output is a draft for human review.

## Instructions to the AI tool

You are reviewing a short research brief. You have access only to the brief and any source notes provided. You do not have access to live data and cannot verify external sources. Do not invent citations or numbers.

Produce a concise, structured review covering each of the following.

### 1. Claim-by-claim source check
- List each substantive claim in the brief.
- For each claim, state whether the source notes appear to support it, partially support it, or do not address it.
- Flag any claim that depends on a source that is not provided.

### 2. Facts vs. interpretation
- Identify sentences that mix description with interpretation.
- Suggest where the brief should separate "the data shows X" from "this might mean Y."

### 3. Uncertainty
- Flag any claim stated with more confidence than the underlying evidence appears to support.
- Suggest more cautious phrasing where appropriate (e.g., "is consistent with," "appears to," "preliminary").

### 4. Overstated conclusions
- Flag conclusions that imply causality where the brief shows only correlation.
- Flag forward-looking statements that are not supported by the source notes.

### 5. Audience and context
- Comment on whether the tone, scope, and level of detail are appropriate for a research-support or policy-support audience.
- Flag any content that would benefit from a clear "this is illustrative" or "this does not represent an official view" disclaimer.

### 6. Human-reviewer checklist
- End with a short list of items a human reviewer should verify before sharing the brief, e.g.:
  - confirm each cited number against the original source,
  - confirm date ranges and units,
  - confirm that interpretive language is appropriately hedged,
  - confirm that the brief does not draw policy conclusions the evidence does not support.

## Reminder

This review is a draft. A human reviewer must read the brief, check sources, and approve or revise the wording before the brief is shared.

when to use it

Community prompt sourced from the open-source GitHub repo yullieyang/llm-research-workflow-assistant (no explicit license). A "Research Brief Review" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.

tags

roleplaycommunitygeneral

source

yullieyang/llm-research-workflow-assistant · no explicit license