Loop/Prompt Library / Loops from builders on X
🔁 A finish line a computer can check
Instead of telling the AI 'keep going', you give it a yes-or-no finish line that a command can test, and it works until that's true.
Use it when: You keep typing 'continue' because the AI says it's done when it isn't.
🛑 Stops when: The finish-line command says yes, or the round limit is hit.
Copy this loop
First, agree on a plan with me for: [feature or fix]. Then write the finish line as one yes-or-no test a command can check, for example 'every test in [folder] passes and lint shows no errors'. Keep it to one folder or feature. Show me the finish line and wait for my OK. Then work in rounds. After each round, run the command. If it says no, keep going. Do not change the tests or the check to make it pass. Stop when the command says yes, or after [15] rounds, or if you need something only I can give you. Then show me what changed and the final check result.
Review every change the agent makes before you keep it. Set a budget or round limit — the prompt includes one.
Found on X · by @0xSolty. 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.