deepen

Silently, before the ask is rewritten, name the field it overlaps — a practice, as "SRE incident response", "investigative journalism" or "clinical pharmacology". Then answer for that field:

1. What would a top professional in it do here that the ask doesn't name?
2. What rawest material would they start from, in place of what has already been digested for the user?
3. Which specific methods and tools would they use?
4. What target would the best in it reach for — one that looks out of reach today, past the field's pass mark and any the ask names?
5. What, other than you or the user, would show the result is right?

Be concrete: name the actual thing, not its category.

Deeper, not bigger: keep the user's goal and scope. Depth is a higher standard for the same outcome — better sources, sharper methods, a harder target, real verification. Drop any answer that adds a deliverable, a feature or a side quest, or that makes the job larger and costlier rather than better. Nothing survives → the ask already meets its field's bar; say so in one line.

Ask first, with AskUserQuestion, about each kept item that spends the user's money or time or changes what ships; the rest stand without asking. Then hand back the field and one line per kept item, as the bar the work will meet:

Bar, as <field> would set it:
- start from: <the rawest material>
- method: <a method or tool>
- aim: <the target, stated as reachable>
- right if: <the outside check>

The ask is still rough — deepening it would mean inventing what the user means → use intent skill. The user wants more options to pick from, not a higher bar for this ask → use consult skill.