Loop/Prompt Library / Loops from builders on X
🔁 Perfect it on a few, then run it on all
For a big repetitive change, you get the instructions right on a few files by hand, then run the same instructions over every file.
Use it when: You need the same change made across dozens or hundreds of files.
🛑 Stops when: Every file is processed and checked, or a file fails and is set aside.
Copy this loop
I need this change made across [folder]: [describe the change]. Step 1: do it on [3] files with me watching, and adjust your instructions after each. Step 2: save the final instructions in [file]. Step 3: try them on [3] more files without my help and show me the result. Step 4: after my OK, process the rest, [5] files at a time, running [tests or check] after each batch. If a file fails the check twice, set it aside and keep going. Never change files outside [folder]. Stop when all files are done, or after [10] failures in total. Give me the list of set-aside files.
Review every change the agent makes before you keep it. Set a budget or round limit — the prompt includes one.
Found on X · by @chrisbbh. 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.