Prompt Screener

Detects biased prompts before you send them to AI — and suggests a neutral reframe.

As of June 2026, Prompt Screener has 9 users and a 5.00/5 rating from 2 reviews in the Workflow & Planning category.

Usersno change0%
9
9
Ratingno change0%
5.00
2 reviews
Reviewsno change0%
2
Version
0.1.1
Manifest V3
90-day change · In the last 90 days this extension 1 version update, changed permissions.

History

6 snapshots

Tracking since May 7, 2026.

10.47.54.6May 7, 2026Jun 13, 2026
View as table
DateUsersRatingReviewsVersion
May 7, 20260.1.0
May 11, 20260.1.0
May 17, 2026105.0020.1.1
May 30, 202685.0020.1.1
Jun 5, 202655.0020.1.1
Jun 13, 202665.0020.1.1
Now95.0020.1.1

Changelog

  • May 11, 2026
    description
    Prompt Screener checks your prompts before you send them to AI and flags patterns that tend to produce one-sided, over-confident, or misleading responses.
    
    When a biased prompt is detected, Prompt Screener pauses the send and shows you why, along with a suggested neutral reframe. You can use the reframe or send your original. Your choice, every time.
    
    What it detects:
    
    Confirmation-seeking — asking AI to validate a conclusion you've already reached
    Loaded presuppositions — treating an unproven claim as established fact
    Authority anchoring — using a previous AI answer to validate itself
    Outcome-leading — asking for evidence in only one direction
    False dichotomies — artificially limiting the options AI considers
    Works on: ChatGPT, Claude, Gemini, and Grok
    
    Your prompts are never stored. Checks are run via a secure proxy and discarded immediately after analysis.
    Prompt Screener checks your prompts before you send them to AI and flags patterns that tend to produce one-sided, over-confident, or misleading responses.
    
    If a biased prompt is detected, Prompt Screener pauses the send and shows you why, along with a suggested neutral reframe. You can use the reframe or send your original.
    
    If no bias is detected, your prompt sends instantly. No interruption.
    
    What it detects:
    Confirmation-seeking — asking AI to validate a conclusion you've already reached
    Loaded presuppositions — treating an unproven claim as established fact
    Authority anchoring — using a previous AI answer to validate itself
    Outcome-leading — asking for evidence in only one direction
    False dichotomies — artificially limiting the options AI considers
    
    Works on: ChatGPT, Claude, Gemini, and Grok
    
    Your prompts are never stored. Checks are run via a secure proxy and discarded immediately after analysis.
  • May 11, 2026
    host_permissions
    https://chatgpt.com/*, https://chat.openai.com/*, https://claude.ai/*, https://gemini.google.com/*, https://grok.com/*, https://prompt-screener-proxy.vercel.app/*
    https://chatgpt.com/*, https://chat.openai.com/*, https://claude.ai/*, https://gemini.google.com/*, https://grok.com/*, https://www.perplexity.ai/*, https://perplexity.ai/*, https://manus.im/*, https://prompt-screener-proxy.vercel.app/*
  • May 11, 2026
    permissions
    (empty)
    storage

Permissions & access

Permissions
storage
Host access
https://chatgpt.com/*, https://chat.openai.com/*, https://claude.ai/*, https://gemini.google.com/*, https://grok.com/*, https://www.perplexity.ai/*, https://perplexity.ai/*, https://manus.im/*, https://prompt-screener-proxy.vercel.app/*

Screenshots

Prompt Screener screenshot 1

About

Prompt Screener checks your prompts before you send them to AI and flags patterns that tend to produce one-sided, over-confident, or misleading responses.

If a biased prompt is detected, Prompt Screener pauses the send and shows you why, along with a suggested neutral reframe. You can use the reframe or send your original.

If no bias is detected, your prompt sends instantly. No interruption.

What it detects:
Confirmation-seeking — asking AI to validate a conclusion you've already reached
Loaded presuppositions — treating an unproven claim as established fact
Authority anchoring — using a previous AI answer to validate itself
Outcome-leading — asking for evidence in only one direction
False dichotomies — artificially limiting the options AI considers

Works on: ChatGPT, Claude, Gemini, and Grok

Your prompts are never stored. Checks are run via a secure proxy and discarded immediately after analysis.

Technical

Version
0.1.1
Manifest
V3
Size
25.29KiB
Min Chrome
88
Languages
1
Featured
No

Metadata

ID
hdooilgdenkeeccfomlkkenhmelobcjm
Developer ID
u391d1324cea453a3df7aa8e0310ad9a5
Developer Email
[email protected]
Created
May 6, 2026
Last Updated (Store)
May 11, 2026
Last Scraped
Jun 13, 2026
Website
Support URL

Data sourced from the Chrome Web Store · last verified Jun 13, 2026.