Loop/Prompt Library / Loops from builders on X
🔁 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.
Use it when: You run an agent that does jobs on a schedule, and things quietly break without you noticing.
🛑 Stops when: Both watchers agree a problem is real and the fix works, or they can't agree and it goes to you.
Copy this loop
You are Watcher A. Read the logs from my agent for the last [24 hours]: [log location]. List every error, failed job, or job that stopped running. For each one, explain the likely cause with the log lines that prove it, and propose one small fix. Then hand your list to Watcher B (a different AI). Watcher B: check each problem and each fix on your own. Say 'agree' or 'disagree' with a reason. Go back and forth at most [3] rounds per problem. Only fixes you both agree on go in a list for me to approve. Anything you can't agree on, mark 'needs a human'. Do not change live settings, delete data, or restart services yourselves. Stop when every problem is either agreed, or marked for me.
Review every change the agent makes before you keep it. Set a budget or round limit — the prompt includes one.
Found on X · by @gkisokay. 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.
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.
Make the AI run your checks
Writes down the checks you normally do by hand, so the AI runs them after every change and must fix any failure before it moves on.