Loop/Prompt Library / Loops from builders on X
🔁 Fix merge conflicts and test again
When two changes clash, the AI untangles them and then runs the full tests again so nothing broken sneaks in.
Use it when: Several changes are waiting to be combined and they keep clashing with each other.
🛑 Stops when: The change passes the full tests after the fix, or it's set aside for a person.
Copy this loop
This change has a merge conflict with the main branch: [branch name]. Read both sides and explain in plain words what each one was trying to do. Fix the conflict so both goals still work. Then run the FULL test suite, not just the parts you touched. If any test fails, fix it and run everything again, at most [3] times. If a test fails sometimes and passes other times, list it as 'flaky' instead of changing it. Don't merge it yourself. Stop when all tests pass and mark it ready for review, or after 3 tries; then explain the clash for a person to decide.
Review every change the agent makes before you keep it. Set a budget or round limit — the prompt includes one.
Found on X · by @nikhil_kadalge. 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.