Dossier

Structured research producing cited, well-edited reports. Two genres: research dossiers that map the state of a topic so the reader draws their own conclusions, and decision dossiers that rank options and recommend one. Optional decision ballots. Works for any domain — tech, travel, health, policy, finance, art.

Pick the genre before researching — it changes every stage below. Getting it wrong is the most expensive error this skill can make: it is invisible until delivery, and it cannot be patched by editing, because the two genres want different documents built from the same facts.

Workflow: SCOPE → GATHER → EVALUATE → SYNTHESIZE → REVIEW → DELIVER

0. Preflight — verify the request is actionable, and settle the genre

Before starting research, confirm all three:

Specific — what to investigate is bounded: a named topic, a concrete comparison, or a clear question. "Look into the X space generically" fails this.
Unambiguous — contested terms are defined by context. If the request uses a word with multiple plausible meanings and context doesn't resolve which, flag it.
Well-understood — you can state the objective back to the operator in 1–2 sentences without hedging. If you'd have to pad with "I think you mean…" or "depending on what you want from this…", you don't understand it yet.

If one or more checks fail, ask before starting. A dossier built on unclear objectives wastes more research time than the clarifying turn costs. Batch questions; offer choices where reasonable; skip the obvious.

Then settle the genre — a fourth, independent check: