Agent teams

The Gauntlet Loop: The Three Line Prompt Where Claude Grades Its Own Work

The three line prompt that makes Claude build, judge and rebuild its own work until a separate critic agent runs out of complaints

6min read
22numbered steps
9sections

What this actually is

The Gauntlet Loop is a way of writing one prompt so that you stop being the person who checks the output. Instead of you reading a result and asking for another pass, the prompt sets up two kinds of agent and lets them argue until the work clears a bar you defined at the start.

It was named and popularised by Matt Shumer, and it spread on the back of one demo he called Claude of Duty: a playable browser shooter, roughly 55,000 lines of Three.js, produced from a single prompt by Claude Opus 5. The demo is the reason people are copying it. The mechanism is the reason it works.

Boris Cherny, who built Claude Code, describes his own day the same way. His words: "My job is to write loops."

Keep reading

Free, for an email.

Unlocks 8 more sections, 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

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
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
Build an Agent Team Where One Agent Attacks the Others' Work
The five prompts that build a team of Claude agents where one agent attacks the work, so what reaches you already survived and you never have to double check it.
4 min