You don't need a hundred operations prompts. You need five you trust and actually reuse.
Start with the smallest version
Write down what you never want as explicitly as what you do. The exclusions often do more work than the instructions.
Reuse the recipe
Treat the first output as a draft, not a verdict. The prompt does the typing; you keep the judgment about what ships.
- Have the model flag what it could not cover
- Keep one ask per prompt — chaining beats cramming
- Name the audience and the single goal before you generate
Cut before you polish
Lock the parts that never change and leave one slot for the variable. Now every run shares a family resemblance instead of starting from scratch.
Ask for the reasoning, then the result. When the model shows its steps you can see exactly where to push back.