Skip to content
TestedFreeEditing

Quick Fact-Check Assistant.

A 30-second accuracy read on any short piece — every claim flagged 🟢/🟠/🔴 with the search term to confirm it.

5 viewsWorks withChatGPTClaudeGeminiGrokDeepSeekPerplexityCopilotAny model
Prompt Locked
01

What this prompt does

  • Flags every empirical claim 🟢/🟠/🔴 in one pass
  • Hands you a search term to confirm each
  • Marks unverifiable claims — never guesses or invents sources
  • Catches accuracy errors before anyone else sees them
02

Why it works

Most 'AI slop' complaints are really accuracy complaints. This catches the obvious errors in one pass, no setup — before anyone else sees them.

03

Tips for this prompt

  • Paste your near-final draft — not a rough first pass.
  • Search every 🔴 and 🟠 term before you ship.
  • Feed it stats-heavy copy — that's where it earns its keep.
  • Add "flag vague sourcing too" for numbers with no source.
  • Trust "Unverifiable" — that's the point, not a miss.
04

How to use the prompt

  1. 1Paste the prompt into your AI assistant.
  2. 2Drop your text into the {{PASTE YOUR TEXT HERE}} slot.
  3. 3Run it.
  4. 4Read the verdict — 🟢, 🟠, or 🔴.
  5. 5Search each flagged claim's term.
  6. 6Fix what needs fixing before you ship.
Tags#fact-checking#editing#quick-check

More like thisEditing prompts

Browse all →
Tested

Make AI writing sound like you

Feed it a draft and get back writing that sounds like you — not an AI model.

ChatGPTContent
2
#136
Tested

Line-by-Line Fact-Checker (IFCN Standards)

Scores every claim True, Context-Missing, False, or Unverifiable — with a red/amber/green publish call and search strings to audit.

Any modelEditing
5
#153
Tested
B - The voice block shown as pasteable text above a rescue prompt, with both marked as one message
Guidea read, not a prompt

Make every reply sound like you [Start Here]

Build a reusable voice block so every rescue reply sounds like you wrote it.

Any modelEmail
18
#086
Tested

Source-Credibility Checker

Paste a URL, outlet, or author and get a cite / corroborate / don't-rely verdict across four credibility axes — plus a stronger primary source to use instead.

ChatGPTResearch
3
#154