Claude Code

The Claude Prompt That Stops It Agreeing With You

AI assistants tend to agree with you, even when you're wrong. Paste this prompt with your plan, resume or essay and get the holes instead of the applause, plus the follow-up that says what would make Claude say no

2min read
3numbered steps
4sections

Ask Claude what it thinks of your plan and you usually get a pat on the back. Anthropic's own researchers found that AI assistants tend to agree with the person they're talking to, even when that person is wrong.

Here's the prompt that flips it, the follow-up, and how to use it on the things you can't afford to get wrong.

The prompt

  1. Paste your plan, resume or essay into Claude.
  2. Under it, type: Don't be nice. Name the three biggest holes in this, and what would make you say no.
  3. Read the holes, fix the one that hurts most, paste the new version and ask the same thing again.

Where it helps

  • A resume: what a hiring manager would reject in the first ten seconds.
  • A business plan: the assumption that breaks it if it's wrong.
  • An essay: the claim a teacher would circle.
  • A decision: the reason you'd regret it in a year.

Free, for your email

Get the full setup steps + the copy-paste setup prompt

The rest of this guide (2 more sections) opens right here. The setup prompt, one paste into Claude that sets it all up, comes by email only. Use your real address.

Unlocks: The catch · Sources

50,000+ follow where these guides come from. No spam, reply “unsubscribe” any time.

Free MCP directory

Make your AI 100x more powerful.

Connect Claude, ChatGPT or Cursor to the apps you already use. Thousands of servers, one click to set up.

Browse free MCP servers

Read next