apw-researchlisted

Use this skill when asked to research, investigate, evaluate options, or understand a problem space *before* committing to an implementation. It is the upstream companion to `apw-implement`: research produces a durable artifact that an implementation can then build on. Follow the steps below strictly and in order. ## Step 1 — Frame the question Before consulting any source, write down what the research must actually answer: - **Core question** — the single decision this research must inform. If you cannot state it in one sentence, the task is not ready to research yet. - **Sub-questions** — the smaller questions the core question breaks down into. - **Why it matters** — what decision or work hangs on the answer. - **Done-criteria** — what "enough to decide" looks like, so the research has a finish line. - **Scope** — what is explicitly in and explicitly out. Present this framing to the user and confirm it before going further. ## Step 2 — Ask questions and choose a strategy (only if needed) If anything about the framing is genuinely ambiguous, ask up to 5 targeted questions that would change how the research is conducted. Skip if the framing is already clear — do not ask for the sake of asking. In this same round, assess **parallelism**: if the sub-questions are independent and non-overlapping — different domains, sources, or codebase areas that can be investigated without one depending on another — tell the user this research could be split across several parallel age