Agent teams

The Second Claude That Reviews the First One's Work

Turn Claude into the brutally honest reviewer that catches what you can't

2min read
1prompts to copy
3numbered steps
5sections

Normal Claude is a yes-man. Ask it to check your work and it says everything you made is amazing, even when it's secretly broken. You believe it, right up until it blows up in your face. Here's how to build a second Claude that tells you the brutal truth instead.

Why self-review never works

The model that made the thing will always defend it. It has an ego to protect, so it tells you what you want to hear. Asking the same Claude "is this good?" is theater. The fix isn't a smarter prompt, it's a second opinion with no skin in the game.

Keep reading

Free, for an email.

Unlocks 4 more sections, 1 copy-paste prompt, the document version, and every other guide on the site, for free. Enter your email once to keep reading.

19,000+ follow where these guides come from.No spam. Unsubscribe in one click.

Read next

Split One Overloaded Claude Into a Seven-Station Line
Turn one overloaded Claude into a 7-station line where nothing ships until the work is checked.
5 min
MetaGPT: One Line In, A Whole Software Company Out
Give it one line and it runs your idea through an actual software company: a product manager, an architect and engineers, each doing their own job before passing the work on
4 min
The Judge Loop: Three Lines That Put Claude's Output On Trial
The three lines that put Claude's own output on trial, the stopping rule that keeps it from looping forever, and an honest account of what it costs
4 min