Loop/Prompt Library / Loops from builders on X
🔁 Diagram maker and critic go back and forth
One AI makes a diagram, another checks it against what it's supposed to show, and they repeat until it's accurate.
Use it when: You need a chart or diagram that correctly explains a process, not just one that looks nice.
🛑 Stops when: The critic finds nothing wrong, or the round limit is hit.
Copy this loop
I need a diagram showing: [paste the description of the process or method]. Step 1: describe the diagram in words first (boxes, arrows, labels) and show me. Step 2: make the image. Step 3: a separate critic AI re-reads my description and lists every mismatch: missing steps, wrong arrows, wrong labels, misleading parts. Step 4: fix them and make a new version. Repeat steps 3-4. Don't add anything that isn't in my description. Stop when the critic finds no mismatches, or after [5] rounds. Show me the final image and the critic's last notes.
Review every change the agent makes before you keep it. Set a budget or round limit — the prompt includes one.
Found on X · by @YoussefHosni951. 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.