Loop/Prompt Library / Loops from builders on X
🔁 A data collector that fixes itself (with a second opinion)
After each scheduled run, a checker grades the collected data. If it failed, an AI patches the collector, and a separate checker must pass the fix before it's used.
Use it when: You collect data from a website on a schedule and it keeps breaking when the site changes.
🛑 Stops when: The fix passes the independent check, or after the try limit it goes to a person.
Copy this loop
My collector gathers [what] from [site] every [day]. Only collect public data and respect the site's rules. After each run, grade the result against these checks: [e.g. 'at least 100 rows', 'no empty prices', 'dates look right']. If it passes, stop. If it fails, look at what changed on the page and patch the collector in a test copy. A separate checker (not you) then runs the patched version and grades it the same way. Only if it passes, swap it in. Try at most [3] patches; if none pass, stop and send me the error and what you tried. Never delete old data.
Review every change the agent makes before you keep it. Set a budget or round limit — the prompt includes one.
Found on X · by @zytedata. 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.