Skip to content

interaction-prompts.md

Latest commit

 

History

History
116 lines (81 loc) · 3 KB

interaction-prompts.md

File metadata and controls

116 lines (81 loc) · 3 KB
Raw

Interaction Prompt Templates

Ready-to-use structured prompts for the interaction convention. Render via the host's AskUserQuestion where available; otherwise as the numbered-list fallback shown here. Always allow a free-text "Other" answer.

Gate (after every phase)

[Gate] {Phase} complete — {one-line outcome}. How do you want to proceed?

  1. Approve (recommended) — accept this phase and continue to {next phase}
  2. Revise — re-run this phase with feedback
  3. Skip {next optional phase} — move on without it (a reason may be required)
  4. Rollback — return to an earlier phase by name
  5. Abort — stop the lifecycle for this feature
  (or type your own answer)

Maps to gates[].decision ∈ approved | revised | skipped | rolled_back | aborted (see policy §2).

Triage / track selection (Theme A, intake)

[Track] This change looks like {classification} ({signals}). Which track?

  1. {recommended track} (recommended) — {why it fits the size}
  2. express — single spec→plan→implement→verify pass (trivial changes)
  3. lite — product-spec → plan → implement → verify (small features)
  4. standard — full lifecycle (research → … → release)
  (or type your own answer)

Persist the choice as feature_mode (express is a first-class mode value) on .forge-status.yml.

Skip reason (policy §3)

[Skip why] You chose to skip {phase}. What's the reason? (required)

  (free text — empty/whitespace is rejected under require_skip_reason: true)

Optional-phase offer

[Run {phase}?] {one-line value of the phase}.

  1. Run {phase} (recommended for {when}) 
  2. Skip — {consequence}
  (or type your own answer)

Next action at phase handoff (Theme I)

[Next step] {Phase} done. What next?

  1. Proceed to Phase {N}: {name} (recommended)
  2. Run optional: {optional phase}
  3. Run a cross-cutting command (sync-verify / change-request / status)
  4. Pause — save state and stop (resume later with /speckit.product-forge.forge)
  (or type your own answer)

Research scope opt-ins (multiSelect)

[Research scope] Which research dimensions should run? (select all that apply)

  - Competitors (default on)
  - UX/UI patterns (default on)
  - Codebase analysis (default on)
  - Tech-stack comparison
  - Metrics / ROI
  (or type your own answer)

Product-spec detail level

[Detail] How detailed should the product spec be?

  1. Standard (recommended) — balanced depth
  2. Concise — minimal, for small/clear features
  3. Exhaustive — deep, for complex/ambiguous features
  (or type your own answer)

Vertical-slice split offer (Theme A)

[Slices] This feature is large. Split into vertical slices that each run
implement→verify independently?

  1. Split into slices (recommended for >10 tasks) — faster feedback, smaller reviews
  2. Keep as one — single implement→verify pass
  (or type your own answer)