Loop/Prompt Library / Loops from builders on X
🔁 Remove the last step you do by hand
Finds the one step you still do by hand that stops a process from running on its own, and keeps trying fixes until that step is automatic.
Use it when: A machine, test, or experiment can run by itself except for one chore you keep having to do.
🛑 Stops when: The process runs [5] full cycles in a row with no help from you, or it runs out of tries.
Copy this loop
Here is my process: [describe it, e.g. 3D printer prints, then I clear the bed]. The step I still do by hand is: [step]. Suggest ways to make that step automatic, safest first. Before trying any idea, tell me what could break or get hurt, and wait for my OK for anything physical. Try one idea at a time. After each try, record what happened. If it fails, adjust one thing and try again. Stop when the process runs [5] cycles in a row without me, or after [15] tries. Then write down the final setup so I can repeat it.
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.