Turn a vague request into a working prompt

Rewrite this request as a prompt that can actually be executed.

**Before writing:**
1. **Restate what I am really asking for** in one sentence. *If you cannot, say so and ask*
2. ***List what you had to guess at, and ask me instead of filling it in.*** **A guess here is why the output is wrong later**

**The rewrite carries:**
- **Task** — one verb, one deliverable
- **Context** — only what changes the output
- **Constraints** — length, format, what to exclude
- **Standard** — what makes the output good, in checkable terms
- **Failure handling** — ***what to do when information is missing.*** **Without this line the model invents**

**Rules:**
- **Do not add "act as a professional X" unless the role changes the answer.** *A role that changes nothing costs tokens and adds nothing*
- **Do not pad with adjectives.** "High-quality", "detailed", "comprehensive" are not instructions
- **Keep my intent.** ***If you find the request itself is a bad idea, say so separately — do not silently fix it***
- Output the prompt as copyable text, **then a short list of what you changed and why**