Loop/Prompt Library / Loops from builders on X
🔁 Same seven steps for every helper
Every AI helper follows the same routine (look, measure, decide, act or suggest, check, write it down, hand off) and leaves a dated report.
Use it when: You run several AI helpers on different parts of a project and can't tell what each one did.
🛑 Stops when: Each helper finishes its check and writes its dated report, or hits the time limit.
Copy this loop
You look after [area, e.g. security or page speed]. Every run, follow these steps in order: 1) Look: read the current state of [area]. 2) Measure: record [1-3 numbers that matter]. 3) Decide: what is the most useful next step? 4) Act or suggest: make one small, safe change, or write a suggestion if the change is risky. 5) Check: measure again and confirm nothing got worse. 6) Write it down: save a report to [reports folder]/[today's date]-[area].md with what you saw, did, and measured. 7) Hand off: list anything the next helper or I need to know. Never merge, deploy, or delete data. Stop after one pass, or after [30] minutes.
Review every change the agent makes before you keep it. Set a budget or round limit — the prompt includes one.
Found on X · by @hambleton_j. 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.