| name | story-step |
| description | Run the next single wave of the story pipeline only, then exit. Manual mode: no in-wave AskUserQuestion, no W3/W7 hard gates, no auto-advance. User reviews deliverables and re-invokes /story-step when ready for the next wave. Trigger: 'one wave', 'single step', 'step through', 'manual mode', 'ํ ๋จ๊ณ๋ง', 'ํ ์จ์ด๋ธ', 'ํ ๋จ๊ณ์ฉ', 'step' |
| argument-hint | |
Run exactly one wave of the 9-wave story pipeline โ the next incomplete one
according to STATE.md โ then exit.
Differences vs /story-execute (auto, full pipeline) and /story-next
(auto, full resume):
- No
AskUserQuestion anywhere (orchestrator OR subagent). Every choice
the wave doc presents as user-facing is auto-resolved with a deterministic
default and logged in the wave's SUMMARY.md.
- No W3/W7 hard gates โ the orchestrator exits after one wave; user
confirms by choosing to invoke
/story-step again.
- No loop to W{N+1} โ exactly one wave per invocation.
Use this when you want to inspect each wave's deliverables before deciding to
continue (vs /story-execute which runs through with two confirmation gates).
<execution_context>
@skills/story-engine/workflows/step.md
@skills/story-engine/workflows/execute-pipeline.md
@skills/story-engine/SKILL.md
</execution_context>
Execute the step workflow from @skills/story-engine/workflows/step.md end-to-end.
Honor the manual-mode contract: no AskUserQuestion, single wave, exit cleanly.