---
name: team-question
description: 'Decomposes a feature into task and question artifacts. Trigger on "shape this idea", "decompose this task", or "/team-question".'
effort: medium
argument-hint: "<ticket id, issue URL, or task description>"
---

Before each dispatch or retry, read [host dispatch](../team/references/15-host-dispatch.md) and supply its resolved installed paths.
Before artifact work, read [artifact schema](../team/references/artifacts.md).
Before handling external values, read [external-data rules](../team/references/external-data.md).

# Team Question — Decompose the Task

Before finalizing prose you author, call the Skill tool with `unslop` and
`writing-prose`, in that order.

Run the QUESTION phase only, then stop. The Question phase decomposes the
user's intent into the artifacts that the rest of the QRSPI pipeline
consumes:

- `1-task.md` — the human's full intent. Read by `design-author` and
  downstream phases that need intent. **Never** read by `researcher` or
  `file-finder` — they only see `2-questions.md`.
- `2-questions.md` — neutral research questions phrased without intent. The
  only file `researcher` and `file-finder` ever read.
- `3-prd.md` — written
  **only when the request is vague, multi-story, cross-cutting, or replaces existing behavior**
  (criteria in the `## Conditional PRD` section of
  `skills/team/playbooks/question.md`, with the template in
  `skills/team/references/prd-template.md`). Referenced
  from `1-task.md`. read downstream by `design-author`.
- `4-repos.md` — written **only when the topic spans more than one
  repository**. Lists each involved repo's slug, absolute path, and
  role. Its presence switches the rest of the pipeline into multi-repo
  mode (one worktree per repo, slice/step `[repo: <slug>]` annotations,
  one PR per repo). See `skills/qrspi-workflow/SKILL.md` for the schema
  and `skills/team/references/multi-repo.md` for the detection rules.

These files live in `docs/plans/<id>/` where `<id>` is either a
ticket-derived slug (`ENG-1234-add-rate-limiting`) or a date-derived slug
(`2026-05-01-add-rate-limiting`).

## Input

`$ARGUMENTS` may be:

- A ticket identifier (e.g. `ENG-1234`) — recorded as `ticketId` on
  `1-task.md`'s frontmatter. The orchestrator does not call any ticketing
  system. The ID is stored for the user's reference.
- An issue URL (e.g. `https://github.com/org/repo/issues/42`) — fetched
  with `gh issue view` (or equivalent) to extract the title and body
  before decomposition.
- Free-form text — treated directly as the feature/task description.

When `$ARGUMENTS` is empty, **discover, do not demand**: ground in repo
context before asking. Read recent `git log` activity and the repo's
`README` / `CLAUDE.md` to propose a likely topic, then use
`AskUserQuestion` with labeled options to fill any genuine gap in intent.
Never bare-stop with a plain "describe it" demand when context is already
available.

## Execution

1. **Resolve the input** to a description:
   - Empty `$ARGUMENTS`: ground in repo context, then ask only for genuine
     gaps, per the **"discover, don't demand"** rule in `## Input`.
   - Ticket-only: ask the user for context, or use any tracker integration
     they have configured to fetch the issue body.
   - Issue URL: run `gh issue view <url> --json title,body` and use the
     title plus body as the description.
   - Free text: use directly.
2. **Derive `<id>`**:
   - If a ticket identifier is present: `<TICKET>-<kebab-topic>` (e.g.,
     `ENG-1234-add-rate-limiting`).
   - Otherwise: `<YYYY-MM-DD>-<kebab-topic>` (e.g.,
     `2026-05-01-add-rate-limiting`).
   - The `<kebab-topic>` is a 2–4 word kebab-case slug derived from the
     description.
3. **Create `docs/plans/<id>/`** if it does not exist.
4. **Resume detection.** If `docs/plans/<id>/1-task.md` already exists,
   re-read it instead of overwriting. If `2-questions.md` is missing, the
   questioner only writes `2-questions.md`.
5. Dispatch the `questioner` agent with the full description and the
   target directory `docs/plans/<id>/`. The agent writes `1-task.md` and
   `2-questions.md`, plus `3-prd.md` when the request meets the PRD criteria
   and `4-repos.md` when it makes sure with the user that the topic spans
   multiple repos.
6. **Stop once `1-task.md` and `2-questions.md` exist on disk** — do not
   continue to RESEARCH. (`3-prd.md` or `4-repos.md` can also exist, neither
   changes the stop condition.)

## When to use

- The idea is vague and you want to see the questioner's framing before
  committing to research.
- You want to review and edit `1-task.md` / `2-questions.md` by hand before
  research begins.
- You want to run multiple research passes against the same task without
  re-decomposing.

Report:

- Path to `1-task.md` and `2-questions.md` (and `3-prd.md` / `4-repos.md` when
  written)
- Topic slug and `<id>`
- Mode: single-repo or multi-repo (with the list of involved repo slugs
  if multi-repo)
- Tell the user: **"Next: run `/team-research docs/plans/<id>/`"**