research-docslisted

# Research Docs UX research runs in four stages — **Plan → Conduct → Analyze → Share**. This skill owns the *documentation artifacts* for three of them: - `plan` — **Plan** (plan + recruit) - `guide` — **Conduct** (moderate + document) - `one-pager` — **Share** (distill + report) **Analyze** — transcript synthesis, quote extraction, theme-finding, confidence checks — is deliberately **out of scope**. It's owned by the team's `customer-insights` skill. Where the workflows meet, this skill prints a handoff pointer. It never replicates analysis logic, re-extracts quotes, or restates the team's quality rules. ## When to use - Explicit invocation: `/research-docs` - "Draft a research plan for <study>" - "Write an interview guide for <study>" - "Make a one-pager from these insights" - User is starting a study (plan), preparing for sessions (guide), or wrapping (one-pager) ## Mode selection If the user's intent is unambiguous from the request, run the matching mode. If unclear, ask which of the three they want — don't guess. ## Operating principles - **Non-intrusive.** Don't impose a folder structure. Write to the current working directory unless the user specifies a path. Don't read from or write to folders owned by other skills. - **No analysis duplication.** The synthesis step is owned by `customer-insights`. This skill never extracts quotes from transcripts, never runs verification passes, never invents themes. - **Reference, don't restate.** When the team's quality rul