Phase 0 — STEP_DONE/TASK_DONE split + owner-follow-up integration smoke

prompts/arbiter-paired-room.md

@@ -34,6 +34,7 @@ You may receive reference opinions from external models appended to your prompt.
 ## Rules
 
 - Base your verdict on evidence (code, test output, logs), not on who said what first
+- When reading owner/reviewer summaries, treat **TASK_DONE** as full task completion, **STEP_DONE** as intermediate progress that should keep the owner flow alive, and **DONE** as a legacy alias for **TASK_DONE**
 - Distinguish reviewer snapshot limits from real product bugs. Reviewer workspaces may intentionally omit heavy artifacts like `node_modules`, `dist`, and `build`; inability to run direct local test/typecheck/build/lint there is not, by itself, a blocker if dedicated verification evidence exists
 - When verification evidence exists from the dedicated verification path, judge that evidence on its merits instead of requiring the reviewer to reproduce the same result from the lightweight reviewer snapshot
 - Your verdict is final for this deadlock cycle — after it, work resumes normally

prompts/claude-paired-room.md

@@ -22,9 +22,11 @@ If you see a materially better design, debugging path, or scoping choice, propos
 
 ## Completion status
 
-**Start your first line** with one of these four statuses. This is required.
+**Start your first line** with one of these six statuses. This is required.
 
-- **DONE** — Approved. The owner's response is correct and complete. Include the evidence
+- **STEP_DONE** — The current step is acceptable, but the original requested task still has remaining work. Send the task back to the owner without escalating to the arbiter
+- **TASK_DONE** — Approved. The owner's work satisfies the full requested task. Include the evidence
+- **DONE** — Legacy alias for **TASK_DONE**. Prefer **TASK_DONE** for new turns
 - **DONE_WITH_CONCERNS** — Approved with concerns. List specific actions the owner must take. If the same concerns repeat for 2+ turns, escalate to BLOCKED
 - **BLOCKED** — Cannot proceed without user decision
 - **NEEDS_CONTEXT** — Missing information from user

prompts/owner-common-paired-room.md

@@ -18,16 +18,19 @@ Challenge the reviewer's reasoning. Point out logical gaps, over-engineering, sc
 
 ## Completion status
 
-**Start your first line** with one of these four statuses. This is required.
+**Start your first line** with one of these six statuses. This is required.
 
-- **DONE** — All steps completed. Include the evidence (test output, build log, diff)
+- **STEP_DONE** — A meaningful intermediate step is complete, but the original task still has remaining work. This keeps the task active and continues the owner flow without reviewer or arbiter intervention
+- **TASK_DONE** — The original requested task is complete. Include the evidence (test output, build log, diff)
+- **DONE** — Legacy alias for **TASK_DONE**. Prefer **TASK_DONE** for new turns
 - **DONE_WITH_CONCERNS** — Completed, but there are issues worth flagging. If the reviewer raises the same concerns again, fix them or escalate to BLOCKED
 - **BLOCKED** — Cannot proceed. State what is stopping you
 - **NEEDS_CONTEXT** — Missing information needed to continue
 
 ### Finalize semantics
 
-- When the reviewer already approved and you are finalizing, **DONE** closes the paired turn
+- When the reviewer already approved and you are finalizing, **TASK_DONE** closes the paired turn
+- In that same finalize step, **STEP_DONE** keeps the task active and resumes the owner flow because the original request still has remaining work
 - In that same finalize step, **DONE_WITH_CONCERNS** does not close the turn — it intentionally reopens review
 - Use **DONE_WITH_CONCERNS** on finalize only when you are explicitly asking the reviewer loop to resume