How do I run the keep closing the gap to the target with an AI agent?

Q: How do I run the keep closing the gap to the target with an AI agent?

Here is the target: [describe it, with a link or picture of the reference]. Here is how we'll judge 'matches': [checklist of what must be the same]. Build a first version. Then each round: compare your version to the target, list the gaps from biggest to smallest, and fix the biggest one. Keep a short log of each round. Only copy things I'm allowed to copy. Stop when every checklist item matches, or after [20] rounds or [time limit], whichever comes first. Then show me the side-by-side comparison and what's still different.

Loop/Prompt Library / Loops from builders on X

🔁 Keep closing the gap to the target

You show the AI exactly what 'great' looks like, and it keeps comparing its work to that target and fixing the biggest gap.

Use it when: You have a big goal with a clear example to match, like 'as good as this app' or 'this exact layout'.

🛑 Stops when: It matches the target on your checklist, or the round or time limit is hit.

Copy this loop

Here is the target: [describe it, with a link or picture of the reference]. Here is how we'll judge 'matches': [checklist of what must be the same]. Build a first version. Then each round: compare your version to the target, list the gaps from biggest to smallest, and fix the biggest one. Keep a short log of each round. Only copy things I'm allowed to copy. Stop when every checklist item matches, or after [20] rounds or [time limit], whichever comes first. Then show me the side-by-side comparison and what's still different.

Review every change the agent makes before you keep it. Set a budget or round limit — the prompt includes one.

Found on X · by @mattshumer_. Prompt rewritten in plain English by FixKit AI.

More loops like this

One useful thing every night

You give the AI a short mission, and each night it picks one task that moves you closer and leaves it for you to check in the morning.

Two watchers keep your agent healthy

Two separate AI helpers read your main agent's logs, find what broke, and must both agree on a fix before it is used.

Weekly skill cleanup

Once a week, the AI grades every saved skill or saved prompt it uses, removes the dead ones, and merges the repeats.

Overnight drafts that learn from results

Each night the AI drafts tomorrow's post from your content plan, and reads your notes about what worked before it writes the next one.

Split the work, keep only what passes

Splits a big job across several AI helpers working side by side, checks each one's work, and keeps only the work that passes.

Practice a website task until it's reliable

The AI tries a website chore, learns from each failed try, and keeps adjusting until it has steps that work every time.