mreq-pro
Installation
SKILL.md
Sharpen the prompt below (or, if none is given, my most recent request). Make it deeper, not bigger. Depth means sharper judgment, better source material, stronger methods and an outside check. It does not mean more deliverables, sections or steps.
Prompt: $ARGUMENTS
When the Prompt line above holds only a placeholder word, not a prompt, my agent did not fill it in: the prompt is the text I gave with this skill.
First, answer these briefly (2-4 lines each):
- The crux. What is the one hard part of this task that decides whether the result is good? What would a top professional do about it that my prompt doesn't ask for? What would a novice get wrong?
- Raw material. What primary source would they start from (raw data, original documents, source code, direct measurement) instead of the summarized or pre-digested version I'm relying on?
- Methods and tools. Which specific methods, tools or standards would they use? Name them. No generic advice.
- Outside check. What evidence, independent of you and me, would show the result is right: a test, a benchmark, a primary source, a measurement? State it as a pass/fail condition defined before the work starts.
If you don't know the field well enough to answer one of these, say so and ask me instead of guessing.