okstra 0.48.0 → 0.49.0

package/docs/superpowers/plans/2026-06-05-compact-markdown-report-tables.md

@@ -0,0 +1,323 @@
+# Compact markdown final-report tables (option X) 구현 계획
+
+> **For agentic workers:** REQUIRED SUB-SKILL: Use superpowers:subagent-driven-development (recommended) or superpowers:executing-plans to implement this plan task-by-task. Steps use checkbox (`- [ ]`) syntax for tracking.
+
+**Goal:** 정본 final-report `.md` 의 narrative 표를 `meta(<br> stack) + prose` 형태로 재구성해 markdown 에디터에서도 핵심 본문이 넓게 읽히게 한다. §5 Clarification 은 평면 유지.
+
+**Architecture:** 변경은 (1) jinja 템플릿 레이아웃 + i18n 라벨 1개, (2) `report_views._inline` 의 `<br>` 보존 + 이제 불필요한 grouping 분기 제거(§5 만 유지)에 한정. `data.json`·report-writer 계약 불변.
+
+**Tech Stack:** jinja2 템플릿, Python 3 (report_views, pytest), JSON i18n. 빌드 `npm run build`.
+
+**설계 근거:** [`docs/superpowers/specs/2026-06-05-compact-markdown-report-tables-design.md`](../specs/2026-06-05-compact-markdown-report-tables-design.md)
+
+**범위 메모:** spec 표 목록(§1, §1.1, §1.2, §3.1, §3.2, §4, Execution Status)에 더해 **§7 Follow-up Tasks** 도 동일 안전 md-merge 대상에 포함한다(코드가 컬럼 파싱하지 않음 — §5 와 달리 안전, 일관성 확보). §5 만 평면 예외.
+
+---
+
+## 파일 구조
+
+| 파일 | 책임 | 작업 |
+|---|---|---|
+| [`templates/reports/final-report.template.md`](../../../templates/reports/final-report.template.md) | §1/§1.1/§1.2/§3.1/§3.2/§4/§7/Exec 표를 meta+prose 로 | Modify |
+| [`templates/reports/i18n/ko.json`](../../../templates/reports/i18n/ko.json) · [`en.json`](../../../templates/reports/i18n/en.json) | `columns.recordMeta` 키 추가 | Modify |
+| [`scripts/okstra_ctl/report_views.py`](../../../scripts/okstra_ctl/report_views.py) | `_inline` `<br>` 보존; generic/Exec/§7 grouping 분기 제거(§5 유지) | Modify |
+| [`tests/test_report_views.py`](../../../tests/test_report_views.py) | `<br>` 보존 + §5 유지 + §1/§3/§4 plain 테스트 | Modify |
+| [`CHANGES.md`](../../../CHANGES.md) | 사용자 영향 항목 | Modify |
+
+작업 순서: 템플릿+i18n(.md 구조) → report_views(`<br>` 보존 + 분기 정리) → CHANGES + 전체 검증 + 실제 재렌더.
+
+---
+
+### Task 1: 템플릿 + i18n — narrative 표를 meta(`<br>`)+prose 로
+
+**Files:**
+- Modify: `templates/reports/i18n/ko.json`, `templates/reports/i18n/en.json`
+- Modify: `templates/reports/final-report.template.md`
+
+- [ ] **Step 1: i18n 에 meta 헤더 키 추가**
+
+`templates/reports/i18n/ko.json` 의 `"columns"` 객체에 추가: `"recordMeta": "항목"`.
+`templates/reports/i18n/en.json` 의 `"columns"` 객체에 추가: `"recordMeta": "Record"`.
+
+- [ ] **Step 2: §1 Summary 표 교체**
+
+`templates/reports/final-report.template.md` 의 §1 표 블록
+```
+| ID | Ticket ID | {{ t("columns.summary") }} | {{ t("columns.source") }} |
+|----|-----------|------------|----------------------------|
+{% for row in summary -%}
+| {{ row.id }} | `{{ row.ticketId }}` | {{ row.summary }} | {{ row.source }} |
+{% endfor %}
+```
+를 다음으로 교체:
+```
+| {{ t("columns.recordMeta") }} | {{ t("columns.summary") }} |
+|--------|------------|
+{% for row in summary -%}
+| **{{ row.id }}**<br>Ticket: `{{ row.ticketId }}`<br>{{ t("columns.source") }}: {{ row.source }} | {{ row.summary }} |
+{% endfor %}
+```
+
+- [ ] **Step 3: §1.1 Consensus 표 교체**
+
+```
+| ID | Ticket ID | Statement | Source items (worker:item) | Evidence (path:line / log / worker report) |
+|----|-----------|-----------|----------------------------|---------------------------------------------|
+{% for row in crossVerification.consensus -%}
+| {{ row.id }} | `{{ row.ticketId }}` | {{ row.statement }} | {{ row.sourceItems | join(', ') }} | {{ row.evidence }} |
+{% endfor %}
+```
+를:
+```
+| {{ t("columns.recordMeta") }} | Statement | Evidence (path:line / log / worker report) |
+|--------|-----------|---------------------------------------------|
+{% for row in crossVerification.consensus -%}
+| **{{ row.id }}**<br>Ticket: `{{ row.ticketId }}`<br>Source items: {{ row.sourceItems | join(', ') }} | {{ row.statement }} | {{ row.evidence }} |
+{% endfor %}
+```
+
+- [ ] **Step 4: §1.2 Differences 표 교체**
+
+```
+| ID | Ticket ID | Disagreement | Workers (position + item) | Evidence |
+|----|-----------|--------------|---------------------------|----------|
+{% for row in crossVerification.differences -%}
+| {{ row.id }} | `{{ row.ticketId }}` | {{ row.disagreement }} | {% for w in row.workersPosition %}{{ w.worker }}:{{ w.itemId }} ({{ w.position }}){% if not loop.last %} / {% endif %}{% endfor %} | {{ row.evidence }} |
+{% endfor %}
+```
+를:
+```
+| {{ t("columns.recordMeta") }} | Disagreement | Evidence |
+|--------|--------------|----------|
+{% for row in crossVerification.differences -%}
+| **{{ row.id }}**<br>Ticket: `{{ row.ticketId }}`<br>Workers: {% for w in row.workersPosition %}{{ w.worker }}:{{ w.itemId }} ({{ w.position }}){% if not loop.last %} / {% endif %}{% endfor %} | {{ row.disagreement }} | {{ row.evidence }} |
+{% endfor %}
+```
+
+- [ ] **Step 5: §3.1 Primary Evidence 표 교체**
+
+```
+| ID | Ticket ID | Evidence | Source items (worker:item) | Source (path:line / log) |
+|----|-----------|----------|----------------------------|---------------------------|
+{% for row in evidence.primary -%}
+| {{ row.id }} | `{{ row.ticketId }}` | {{ row.evidence }} | {{ row.sourceItems | join(', ') }} | {{ row.source }} |
+{% endfor %}
+```
+를:
+```
+| {{ t("columns.recordMeta") }} | Evidence |
+|--------|----------|
+{% for row in evidence.primary -%}
+| **{{ row.id }}**<br>Ticket: `{{ row.ticketId }}`<br>Source items: {{ row.sourceItems | join(', ') }}<br>Source: {{ row.source }} | {{ row.evidence }} |
+{% endfor %}
+```
+
+- [ ] **Step 6: §3.2 Secondary 표 교체**
+
+```
+| ID | Ticket ID | Hypothesis or supporting evidence | Source / confidence |
+|----|-----------|-----------------------------------|---------------------|
+{% for row in evidence.secondary -%}
+| {{ row.id }} | `{{ row.ticketId }}` | {{ row.hypothesis }} | {{ row.confidence }} |
+{% endfor %}
+```
+를:
+```
+| {{ t("columns.recordMeta") }} | Hypothesis or supporting evidence | Source / confidence |
+|--------|-----------------------------------|---------------------|
+{% for row in evidence.secondary -%}
+| **{{ row.id }}**<br>Ticket: `{{ row.ticketId }}` | {{ row.hypothesis }} | {{ row.confidence }} |
+{% endfor %}
+```
+
+- [ ] **Step 7: §4 Risks 표 교체**
+
+```
+| ID | Ticket ID | Item | Risk if ignored | Mitigation Owner |
+|----|-----------|------|-----------------|------------------|
+{% for row in missingInformation -%}
+| {{ row.id }} | `{{ row.ticketId }}` | {{ row.item }} | {{ row.risk }} | {{ row.owner }} |
+```
+의 헤더/구분/row 3줄을:
+```
+| {{ t("columns.recordMeta") }} | Item | Risk if ignored | Mitigation Owner |
+|--------|------|-----------------|------------------|
+{% for row in missingInformation -%}
+| **{{ row.id }}**<br>Ticket: `{{ row.ticketId }}` | {{ row.item }} | {{ row.risk }} | {{ row.owner }} |
+```
+(`{% endfor %}` 이후는 그대로.)
+
+- [ ] **Step 8: Execution Status 표 교체**
+
+```
+| Agent | Role | Model | Status | {{ t("columns.rawTokens") }} | {{ t("columns.billableTokens") }} | {{ t("columns.cost") }} | Duration | Summary of Key Findings |
+|-------|------|-------|--------|-----------|-----------|------------|----------|-------------------------|
+{% for row in executionStatus -%}
+| {{ row.agent }} | {{ row.role }} | {{ row.model }} | {{ row.status }} | {{ row.totalTokens | format_int }}{% if row.cliTotalTokens %} (CLI: {{ row.cliTotalTokens | format_int }}){% endif %} | {{ row.billableTokens | format_int }} | {{ row.costUsd | format_usd }}{% if row.cliCostUsd %} (+ CLI {{ row.cliCostUsd | format_usd }}){% endif %} | {{ row.durationMs | format_duration_ms }} | {{ row.summary }} |
+{% endfor %}
+```
+를:
+```
+| {{ t("columns.recordMeta") }} | Summary of Key Findings |
+|--------|-------------------------|
+{% for row in executionStatus -%}
+| **{{ row.agent }}**<br>Role: {{ row.role }}<br>Model: {{ row.model }}<br>Status: {{ row.status }}<br>{{ t("columns.rawTokens") }}: {{ row.totalTokens | format_int }}{% if row.cliTotalTokens %} (CLI: {{ row.cliTotalTokens | format_int }}){% endif %}<br>{{ t("columns.billableTokens") }}: {{ row.billableTokens | format_int }}<br>{{ t("columns.cost") }}: {{ row.costUsd | format_usd }}{% if row.cliCostUsd %} (+ CLI {{ row.cliCostUsd | format_usd }}){% endif %}<br>Duration: {{ row.durationMs | format_duration_ms }} | {{ row.summary }} |
+{% endfor %}
+```
+
+- [ ] **Step 9: §7 Follow-up Tasks 표 교체**
+
+§7 표를 찾아(헤더에 `Title` / `Scope` / `Reason`), 짧은 컬럼(ID, Ticket ID, Origin, New Task ID, Suggested task-type, Priority, Auto-spawn 등)을 meta 셀에 `<br>` stack 하고 Title/Scope/Reason 을 prose 컬럼으로 둔다. 헤더를 `| {{ t("columns.recordMeta") }} | Title | Scope (files/areas) | Reason / Why deferred |` 로, row 의 meta 셀은 `**{{ row.id }}**<br>Ticket: \`{{ row.ticketId }}\`<br>Origin: {{ row.origin }}<br>New Task ID: {{ row.newTaskId }}<br>Type: {{ row.suggestedTaskType }}<br>Priority: {{ row.priority }}<br>Auto-spawn: {{ row.autoSpawn }}` 로 구성(현 row 의 필드명을 그대로 사용 — 먼저 현 §7 row 의 jinja 필드명을 읽어 정확히 매핑할 것). 빈 상태 분기 유지.
+
+- [ ] **Step 10: §5 는 건드리지 않음 (확인)**
+
+§5 Clarification Items 의 8-컬럼 표(`| ID | Ticket ID | Kind | Statement | Expected form | Blocks | Status | User input |`)는 **변경하지 않는다**. grep 으로 §5 표가 8-컬럼 그대로인지 확인.
+
+- [ ] **Step 11: 렌더 스모크 + 커밋**
+
+기존 렌더 테스트로 회귀 확인 후, 픽스처 data.json 으로 렌더해 §1/§1.1/§1.2/§3.1/§3.2/§4/§7/Exec 가 `recordMeta` 헤더 + `<br>` 를 포함하고 §5 가 8-컬럼인지 확인:
+```bash
+python3 -m pytest tests/test_render_final_report.py tests/test_template_full_render_both_langs.py -q
+```
+Expected: PASS (템플릿 문법 OK). 실패 시 jinja 문법/필드명 교정.
+
+```bash
+git add templates/reports/final-report.template.md templates/reports/i18n/ko.json templates/reports/i18n/en.json
+git commit -m "feat(report-template): compact narrative tables as meta(<br>)+prose; §5 stays flat"
+```
+
+---
+
+### Task 2: report_views — `<br>` 보존 + grouping 분기 정리
+
+**Files:**
+- Modify: `scripts/okstra_ctl/report_views.py`
+- Modify: `tests/test_report_views.py`
+
+- [ ] **Step 1: 실패 테스트 작성 (`<br>` 보존 + §1 plain)**
+
+`tests/test_report_views.py` 끝에 추가:
+```python
+def test_inline_preserves_br_tags():
+    from okstra_ctl.report_views import _inline  # noqa: PLC0415
+    out = _inline("**C-1**<br>Ticket: `DEV-1`<br>Source items: claude:F-001")
+    assert "<br>" in out
+    assert "&lt;br&gt;" not in out
+    assert "<strong>C-1</strong>" in out
+    assert "<code>DEV-1</code>" in out
+
+
+def test_summary_meta_table_renders_plain_with_br():
+    # §1 Summary is now pre-merged in the .md (meta col + Summary col); the
+    # grouped-table branch must NOT fire (no separate `Ticket ID` column) and
+    # the <br> in the meta cell survives.
+    html_out = _emit(
+        "| 항목 | 한 줄 요약 |",
+        "| **P-001**<br>Ticket: `DEV-9184`<br>출처: task-brief.md:19 | " + ("on-the-fly 계산 전환 핵심 변경 " * 4) + " |",
+        section="1. Summary of the Problem or Verification Target",
+    )
+    assert 'class="grouped-table"' not in html_out
+    assert "<br>" in html_out
+```
+
+- [ ] **Step 2: 실패 확인**
+
+Run: `python3 -m pytest tests/test_report_views.py -q -k "inline_preserves_br or summary_meta_table"`
+Expected: FAIL — `_inline` escapes `<br>`; §1 still hits the generic grouping branch (grouped-table present).
+
+- [ ] **Step 3: `_inline` 가 `<br>` 보존**
+
+`scripts/okstra_ctl/report_views.py` `_inline` 의 `return out` 직전에 추가:
+```python
+    # Preserve explicit <br> line breaks used inside compact meta cells (the
+    # markdown source intentionally stacks short fields with <br>). They are
+    # escaped to &lt;br&gt; by html.escape above; restore the tag.
+    out = out.replace("&lt;br&gt;", "<br>").replace("&lt;br/&gt;", "<br>").replace("&lt;br /&gt;", "<br>")
+```
+
+- [ ] **Step 4: generic/Exec/§7 grouping 분기 제거 (§5 유지)**
+
+`_grouped_table_spec` 에서 다음을 **삭제**:
+- "Execution Status by Agent" 분기 (`if len(norm) >= 3 and norm[0] == "Agent" ...`).
+- "§7 Follow-up Tasks" 분기 (`if any("Follow-up Tasks" in h ...)`).
+- generic "Ticket ID" 분기 (`if any(h == "Ticket ID" for h in norm): ...`) + 그 헬퍼 `_is_force_meta`, `_column_is_wide`, 상수 `_WIDE_PROSE_TOKENS`, `_FORCE_META_TOKENS`, `_WIDE_CONTENT_THRESHOLD`, `_FOLLOWUP_WIDE_PREFIXES` (이제 미사용).
+- **유지:** §5 Clarification 분기(Expected form/Statement/User input wide) + signature 의 `rows` 파라미터(호출부 호환). `rows` 가 더는 분기 로직에 안 쓰이면, 호출부 `_grouped_table_spec(header_cells, rows, section_path)` 도 `_grouped_table_spec(header_cells, section_path)` 로 되돌리고 시그니처에서 `rows` 제거(미사용 인자 정리).
+
+docstring 을 "§5 Clarification Items 만 grouped (interactive form). 나머지 narrative 표는 템플릿에서 이미 compact 하게 렌더되므로 여기서 grouping 하지 않는다." 로 갱신.
+
+- [ ] **Step 5: 통과 확인 + 기존 grouped 테스트 정리**
+
+Run: `python3 -m pytest tests/test_report_views.py -q`
+Expected: 신규 2개 PASS. **단, Task 직전(2343e30)에 추가한 §1/§3/§4 grouped 테스트**(`test_summary_table_groups_short_cols_and_widens_prose`, `test_risks_table_widens_item_risk_mitigation`)는 이제 의도가 바뀌었으므로 제거하거나 plain+`<br>` 기대로 갱신한다. `test_clarification_expected_form_is_wide_not_meta` 는 **유지**(§5 grouping 살아있음). 모든 report_views 테스트 PASS 확인.
+
+- [ ] **Step 6: 빌드 + 커밋**
+
+```bash
+npm run build && bash validators/validate-workflow.sh
+git add scripts/okstra_ctl/report_views.py tests/test_report_views.py runtime/
+git commit -m "refactor(report-views): preserve <br>; group only §5 (narrative tables compact in md)"
+```
+
+---
+
+### Task 3: CHANGES + 전체 검증 + 실제 재렌더
+
+**Files:**
+- Modify: `CHANGES.md`
+
+- [ ] **Step 1: CHANGES 항목 추가**
+
+`## 2026-06-05` 아래(직전 report-views 항목 근처)에 삽입:
+```markdown
+### feat(report-template): 정본 final-report `.md` 표를 compact 하게 (meta + prose)
+
+- markdown 표는 컬럼 병합이 안 돼 ID·Ticket·Source 같은 짧은 코드 컬럼이 칸을 차지하면 요약·근거·이견·위험 같은 긴 본문이 좁아져 뭉개졌다(에디터에서 한 글자/줄). 이제 §1 Summary·§1.1 Consensus·§1.2 Differences·§3.1/§3.2 Evidence·§4 Risks·§7 Follow-up·Execution Status 표를 **짧은 코드 필드는 `<br>` 로 한 meta 셀에 stack + 긴 본문은 별도 컬럼**으로 렌더한다. §5 Clarification 은 carry-in 파서·validator 8-컬럼 계약 때문에 평면 유지(§5 compact 는 HTML view grouping 담당). `data.json`·report-writer 계약은 불변(템플릿 레이아웃만). HTML self-contained view 도 `_inline` 이 `<br>` 를 보존해 동일하게 compact 하게 보인다.
+- 사용자 영향: 다음 release + `npx -y okstra@latest install` 후 적용. 이제 final-report 를 어떤 markdown 에디터로 열어도 핵심 본문이 넓게 읽힌다. `.md`↔HTML 레이아웃이 일관된다.
+```
+
+- [ ] **Step 2: 전체 검증**
+
+```bash
+npm run build
+python3 -m pytest tests/ -q
+bash validators/validate-workflow.sh
+node bin/okstra --version
+```
+ALL must pass (worktree 격리 flake 예외는 단독 재실행 확인). 그 외 실패 → STOP/BLOCKED.
+
+- [ ] **Step 3: 실제 재렌더 검증 (BLOCKING — 육안)**
+
+기존 사용자 리포트를 /tmp 복사본으로 재렌더하고 구조 확인:
+```bash
+SRC="/Volumes/Workspaces/workspace/projects/FontsNinja/app/fontradar-v2-api/.okstra/tasks/calcule-des-prix-1-1/dev-9184/runs/requirements-discovery/reports/final-report-requirements-discovery-001.data.json"
+```
+NOTE: HTML view 는 `.md` 에서 파생되므로, **새 템플릿으로 `.md` 를 다시 렌더**해야 한다. data.json → md 렌더는 `scripts/okstra-render-final-report.py`(또는 report-writer 경로) 사용. 해당 CLI 의 인자를 `--help` 로 확인 후, /tmp 에 새 `.md` 를 렌더하고:
+- 새 `.md` 의 §1/§1.1/§1.2/§3.1/§3.2/§4/§7/Exec 가 `**ID**<br>Ticket: …` meta 셀 + prose 컬럼인지,
+- §5 가 여전히 `| ID | Ticket ID | Kind | Statement | Expected form | Blocks | Status | User input |` 8-컬럼인지,
+- 그 `.md` 로 `okstra-render-report-views.py` 를 돌려 HTML 의 meta 셀이 `<br>` 줄바꿈으로 보이는지(literal `&lt;br&gt;` 아님),
+- `python3 -c "import sys; sys.path.insert(0,'scripts'); from okstra_ctl.clarification_items import parse_clarification_items; print(len(parse_clarification_items(open('<new-md>').read()) or []))"` 로 §5 carry-in 파서가 여전히 행을 파싱하는지
+확인. /tmp 산출물은 정리(사용자 프로젝트 파일은 건드리지 않음).
+
+- [ ] **Step 4: 최종 커밋**
+
+```bash
+git add CHANGES.md
+git commit -m "docs(changes): log compact markdown final-report tables"
+```
+
+---
+
+## Self-Review (작성자 체크리스트)
+
+**1. Spec coverage**
+- §2.1 대상 표 7종 + §7 → Task 1 Step 2–9.
+- §2.1 §5 평면 유지 → Task 1 Step 10 + 미변경.
+- §2.2 meta `<br>` 포맷 + i18n → Task 1 Step 1–9.
+- §2.3 `_inline` `<br>` 보존 → Task 2 Step 3.
+- §2.3 generic/Exec 분기 제거, §5 유지 → Task 2 Step 4(§7 도 제거).
+- §2.4 계약 불변 → Task 1(템플릿만), §5 미변경.
+- §4 검증(실제 재렌더 + §5 파서) → Task 3 Step 2–3.
+
+**2. Placeholder scan:** §7(Step 9)은 현 row 필드명을 "먼저 읽어 매핑"하라고 명시 — 구현자가 실제 필드명 확인 후 작성(추측 금지). 그 외 모든 jinja/코드 블록은 실제 내용. TBD 없음.
+
+**3. 식별자 일관성:** `columns.recordMeta`(i18n, Task1) ↔ 템플릿 헤더 사용 동일. `_inline`(Task2) ↔ report_views. `_grouped_table_spec` 시그니처에서 `rows` 제거 시 호출부도 함께(Task2 Step4 명시).

package/docs/superpowers/specs/2026-06-05-compact-markdown-report-tables-design.md

@@ -0,0 +1,87 @@
+# 정본 final-report `.md` 표를 compact 하게 (옵션 X) — 설계
+
+- 작성일: 2026-06-05
+- 범위: final-report 정본 `.md` 의 narrative 표(§1 Summary, §1.1 Consensus, §1.2 Differences, §3.1 Primary Evidence, §3.2 Secondary, §4 Risks, Execution Status by Agent)를 **짧은 코드 컬럼을 `<br>` 로 한 셀에 stack 한 meta 컬럼 + 긴 prose 컬럼은 별도 컬럼** 으로 재구성해, markdown 에디터에서도 핵심 본문(요약·근거·이견·위험)이 넓게 읽히도록 한다. 이는 [`templates/reports/final-report.template.md`](../../../templates/reports/final-report.template.md) 의 jinja 레이아웃만 바꾸며, `data.json` 스키마·report-writer 계약은 불변이다.
+- 비범위
+  - **§5 Clarification Items 는 평면 8-컬럼 유지** — [`scripts/okstra_ctl/clarification_items.py`](../../../scripts/okstra_ctl/clarification_items.py) 가 `--resume-clarification` carry-in 을 위해 `|` 8-컬럼으로 파싱하고 validator 가 8-컬럼 스키마를 BLOCKING 으로 강제. §5 의 compact 는 HTML view 의 기존 grouping(이미 Expected form wide 까지 교정됨)이 담당한다.
+  - `data.json` 스키마·report-writer worker 의 출력 계약·convergence 상태 변경 없음 (같은 필드를 템플릿이 다르게 배치할 뿐).
+  - implementation-planning §4.5 deliverable 표(Stage Map / Stepwise 등)는 비대상 — validator 가 그 컬럼/헤딩을 substring 검사하므로 손대지 않는다.
+- 관계: 직전 작업(브랜치 `fix/report-table-grouping`, 커밋 `2343e30`)은 **HTML view 에서만** §1/§3/§4 를 grouped 로 만들었다. 본 설계(X)는 §1/§3/§4 를 `.md` 자체에서 compact 하게 만들어 그 HTML-only 접근을 **대체**한다 — 해당 표들은 더 이상 별도 `Ticket ID` 컬럼이 없어 report_views 의 generic grouping 분기가 발동하지 않으므로, 그 분기를 정리한다. §5 grouping(+Expected form wide) 은 유지된다.
+
+## 1. 동기
+
+사용자는 final-report 를 **`.md` 파일로 markdown 에디터에서 읽는다.** markdown 표는 colspan 이 없어, `ID·Ticket ID·Source·Kind·Status` 같은 짧은 코드 컬럼이 각각 한 칸씩 차지하면 정작 긴 prose 컬럼(요약·Statement·Evidence·Disagreement·Item·Risk)이 좁아져 세로로 한 글자씩 뭉개진다(실측: §1 Summary 의 "한 줄 요약"이 1글자/줄). HTML self-contained view 는 grouping 으로 해결되지만 `.md` 를 읽는 사용자에겐 닿지 않는다. 따라서 `.md` 자체에서 짧은 컬럼을 `<br>` 로 한 셀에 모아 컬럼 수를 줄이고 prose 에 폭을 준다.
+
+## 2. 핵심 설계
+
+### 2.1 대상 표와 meta/wide 분해
+
+각 표를 `[meta 컬럼] + [prose 컬럼들]` 로 재구성한다(meta = 짧은 코드 필드를 `<br>` stack):
+
+| 표 | meta 컬럼(한 셀에 `<br>` stack) | 별도 prose 컬럼 |
+|---|---|---|
+| §1 Summary | ID, Ticket ID, 출처 | 한 줄 요약 |
+| §1.1 Consensus | ID, Ticket ID, Source items | Statement, Evidence |
+| §1.2 Differences | ID, Ticket ID, Workers(position+item) | Disagreement, Evidence |
+| §3.1 Primary Evidence | ID, Ticket ID, Source items, Source(path:line) | Evidence |
+| §3.2 Secondary | ID, Ticket ID | Hypothesis or supporting evidence, Source / confidence |
+| §4 Risks | ID, Ticket ID | Item, Risk if ignored, Mitigation Owner |
+| Execution Status | Agent, Role, Model, Status, raw/billable tokens, cost, Duration | Summary of Key Findings |
+
+빈 상태(empty) 분기는 현행 그대로 유지(`emptyState.*`).
+
+### 2.2 meta 셀 포맷 (`<br>` stack, i18n)
+
+meta 셀은 headline(주 식별자) + `<br>` 로 이어지는 `라벨: 값` 들로 구성한다. 예 §1.1 한 row:
+
+```
+**C-1**<br>{ticketLabel}: `DEV-9184`<br>{sourceLabel}: claude:F-001, codex:1.1
+```
+
+- headline 은 ID(또는 Agent/Step)를 `**굵게**`.
+- 라벨은 기존 i18n 키를 재사용(`columns.ticketId`, `columns.source`, …); 없으면 신규 키 추가(예 `columns.recordMeta` = meta 컬럼 헤더 "항목"/"Record").
+- 값에 코드성인 것(ticket 등)은 기존처럼 백틱 유지.
+- meta 컬럼 헤더는 짧은 라벨(예 "항목" / "Record") — 신규 i18n 키.
+
+### 2.3 report_views 정합
+
+1. **`_inline` 가 `<br>` 를 보존**: 현재 `html.escape` 가 `<br>`→`&lt;br&gt;` 로 깨뜨린다. escape 후 `&lt;br&gt;` / `&lt;br/&gt;` / `&lt;br /&gt;` 를 `<br>` 로 복원(bold/code/link 복원과 동일 패턴). 이로써 HTML view 도 meta 셀이 줄바꿈으로 보인다.
+2. **generic Ticket-ID grouping 분기 제거**: §1/§3/§4 는 이제 `Ticket ID` 단독 컬럼이 없어 그 분기가 발동하지 않는다 — commit `2343e30` 가 추가한 generic 분기(+ 관련 헬퍼·테스트)를 정리한다. Execution Status 도 `.md` 에서 merged 되므로 그 explicit 분기 제거. **§5 Clarification grouping(+Expected form wide)만 남긴다**(§5 는 `.md` 평면 유지 → HTML 에서 grouping).
+3. **plain-table 폭 보강**: merged meta 컬럼은 좁게, prose 컬럼은 넓게 나오도록 plain 경로의 컬럼 폭 처리를 점검(필요 시 meta 컬럼에 narrow, prose 에 min-width).
+
+### 2.4 계약 영향
+
+- `data.json`·report-writer 계약·convergence 상태: **불변**. 템플릿이 같은 데이터를 다르게 배치.
+- §5 파서(`clarification_items.py`)·§5 8-컬럼 validator: **불변**(§5 평면 유지).
+- §1/§3/§4 컬럼 헤더를 substring 검사하는 validator/테스트는 없음(확인됨); 구현 단계에서 grep + 전체 테스트로 재확인.
+- carry-in 으로 다음 run 에 들어가는 `.md` 의 §1/§3/§4 는 LLM 이 컨텍스트로 읽을 뿐 코드가 컬럼 파싱하지 않으므로 `<br>` stack 도 안전.
+
+## 3. 변경 파일
+
+1. [`templates/reports/final-report.template.md`](../../../templates/reports/final-report.template.md) — §1/§1.1/§1.2/§3.1/§3.2/§4 + Execution Status 표를 meta(`<br>` stack) + prose 형태로 재작성.
+2. [`templates/reports/report.i18n.*`](../../../templates/reports/) (또는 i18n SOT) — meta 컬럼 헤더 + 필요한 라벨 키 추가(ko/en).
+3. [`scripts/okstra_ctl/report_views.py`](../../../scripts/okstra_ctl/report_views.py) — `_inline` `<br>` 보존; generic Ticket-ID grouping 분기 + Execution Status 분기 제거(§5 grouping 유지); plain 폭 보강.
+4. [`tests/test_report_views.py`](../../../tests/test_report_views.py) — `<br>` 보존 테스트; §1/§3/§4 가 더는 grouped 분기 안 타고 `<br>` 가 보존되는지; §5 는 여전히 grouped + Expected form wide.
+5. (필요 시) [`tests/test_render_*`](../../../tests/) — 템플릿 렌더 결과의 표 구조 스냅샷/스모크.
+6. [`CHANGES.md`](../../../CHANGES.md) — 사용자 영향 항목.
+
+## 4. Enforcement / 검증
+
+- 단위: `test_report_views.py` 로 `_inline` `<br>` 보존 + §5 grouping 유지를 잠금.
+- 템플릿 렌더: 실제 `data.json`(또는 픽스처)로 렌더해 §1/§3/§4/Exec 표가 meta+prose 형태로 나오는지 + §5 가 평면 8-컬럼인지 확인.
+- **실제 재렌더 검증(BLOCKING)**: 기존 사용자 리포트(또는 픽스처)를 렌더해 `.md` 와 HTML view 양쪽이 compact 하게 나오고 §5 carry-in 파서가 여전히 8-컬럼을 파싱하는지 실행 확인.
+- `python3 -m pytest tests/` + `bash validators/validate-workflow.sh` + `npm run build` 통과.
+
+## 5. 트레이드오프 / 리스크
+
+- **트레이드오프:** §1/§3/§4 의 HTML view 가 grp-meta "key: value" 폴리시 대신 `<br>` stack 형태가 된다(살짝 덜 꾸며짐). 대신 `.md`↔HTML 레이아웃이 일관되고 `.md` 자체가 어떤 에디터에서도 읽기 쉽다(사용자 목표).
+- **리스크 — 전 task-type 영향:** 템플릿이 공유라 모든 phase 의 final-report `.md` 구조가 바뀐다. 기계 파싱은 §5 뿐이라 안전하나, 구현 시 전체 테스트 + 실제 렌더로 회귀 확인.
+- **리스크 — `<br>` 미지원 에디터:** 드물게 표 셀 `<br>` 를 literal 로 보이는 렌더러가 있을 수 있음. 주류(GitHub/Obsidian/Typora/VS Code)는 지원. 정본 가독성 목표상 수용.
+
+## 6. 수용 기준
+
+1. final-report `.md` 의 §1/§1.1/§1.2/§3.1/§3.2/§4 + Execution Status 가 meta(`<br>` stack) + prose 컬럼 형태로 렌더된다.
+2. §5 Clarification 은 평면 8-컬럼 유지, carry-in 파서·validator 통과.
+3. HTML view 가 meta 셀의 `<br>` 를 줄바꿈으로 보여준다(`_inline` 보존).
+4. report_views 의 §1/§3/§4 generic grouping·Execution Status 분기는 제거되고 §5 grouping(+Expected form wide)은 유지.
+5. `python3 -m pytest tests/` + validator + build 통과 + 실제 재렌더 육안 확인.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "okstra",
-  "version": "0.48.0",
+  "version": "0.49.0",
   "description": "Multi-agent cross-verification orchestrator runtime + Claude Code skills.",
   "license": "MIT",
   "author": "devonshin",

package/runtime/BUILD.json

@@ -1,5 +1,5 @@
 {
-  "package": "0.48.0",
-  "builtAt": "2026-06-04T17:18:36.379Z",
+  "package": "0.49.0",
+  "builtAt": "2026-06-05T09:11:13.878Z",
   "repoRoot": "/home/runner/work/okstra/okstra"
 }

package/runtime/bin/lib/okstra/tmux-pane.sh

@@ -0,0 +1,40 @@
+#!/usr/bin/env bash
+# Resolve the tmux pane that the CURRENT process actually runs in.
+#
+# Why this exists: Claude Code's Bash tool strips $TMUX and $TMUX_PANE, so a
+# bare `tmux display-message -p '#{pane_id}'` does NOT return the caller's pane
+# — it returns the pane of the most-recently-active tmux *client*, which (when
+# the user has several attached sessions) is frequently a DIFFERENT session
+# than the one the okstra run lives in. Earlier trace-pane fixes all trusted
+# `display-message` and therefore mis-placed (or dropped) the tail pane.
+#
+# This resolver instead walks the process's own ancestor PIDs and matches them
+# against the tmux server's pane_pids. That is deterministic and correct
+# regardless of $TMUX/$TMUX_PANE or which client is active: when the process is
+# a descendant of a tmux pane's shell it finds exactly that pane; when it is not
+# inside any tmux pane (e.g. Claude launched from the macOS GUI app) no ancestor
+# matches and the function prints nothing.
+#
+# Usage: pane="$(okstra_resolve_caller_pane)"   # empty => not in a tmux pane
+# Optional arg: a starting PID (defaults to $$) — used by the regression test.
+# bash 3.2 safe (no associative arrays).
+okstra_resolve_caller_pane() {
+  command -v tmux >/dev/null 2>&1 || return 0
+  local panes
+  panes="$(tmux list-panes -a -F '#{pane_pid} #{pane_id}' 2>/dev/null)" || return 0
+  [ -n "$panes" ] || return 0
+
+  local pid="${1:-$$}"
+  local depth=0
+  local hit
+  while [ -n "$pid" ] && [ "$pid" != "0" ] && [ "$depth" -lt 16 ]; do
+    hit="$(printf '%s\n' "$panes" | awk -v p="$pid" '$1==p {print $2; exit}')"
+    if [ -n "$hit" ]; then
+      printf '%s\n' "$hit"
+      return 0
+    fi
+    pid="$(ps -o ppid= -p "$pid" 2>/dev/null | tr -d ' ')"
+    depth=$((depth + 1))
+  done
+  return 0
+}

package/runtime/bin/okstra-codex-exec.sh

@@ -183,36 +183,32 @@ status_path="${prompt_path%.md}.status.json"
 [[ "$status_path" == "$prompt_path" ]] && status_path="${prompt_path}.status.json"
 started_ts=$(date +%s)
 script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd -P)"
+# Trace-pane caller resolution helper (okstra_resolve_caller_pane). The lib dir
+# is a bin-sibling in both repo (scripts/lib/...) and installed
+# (~/.okstra/bin/lib/...) layouts; degrade silently if absent.
+[ -r "$script_dir/lib/okstra/tmux-pane.sh" ] && . "$script_dir/lib/okstra/tmux-pane.sh"
 python3 "$script_dir/okstra-wrapper-status.py" \
   init "$status_path" "$(basename "$0")" "$role" "$$" "$started_ts" "$log_path" \
   >>"$log_path" 2>&1 || true
 
 # Derive the okstra run dir from the prompt path. paths.py is the SSOT:
 # dispatched prompts live at `<RUN_DIR>/prompts/<cli>-worker-prompt<NNN>.md`,
-# so the run dir is two levels up. Used to (a) read the lead pane the lead
-# recorded in its own foreground pane and (b) tag the trace pane so cleanup
-# can find exactly this run's panes without any tmux env var. Empty if the
+# so the run dir is two levels up. Used to tag the trace pane so cleanup can
+# find exactly this run's panes without any tmux env var. Empty if the
 # derivation fails — every dependent step below then degrades to a no-op.
 run_dir="$(cd "$(dirname "$prompt_path")/.." 2>/dev/null && pwd -P || true)"
-lead_pane_file="${run_dir:+$run_dir/state/lead-pane.id}"
 
-# Resolve the pane to anchor the trace split to. Claude Code's Bash tool now
-# strips BOTH `$TMUX` and `$TMUX_PANE`, and this wrapper frequently runs
-# backgrounded — so the bare active-pane probe can land on whatever pane the
-# user happens to be looking at now, not Claude's. Prefer the lead pane the
-# lead captured ONCE in its own foreground pane (reliable, see
-# `_common-contract.md`); fall back to `$TMUX_PANE`, then the active-pane
-# probe. A stale recorded id (pane since closed) is rejected via a liveness
-# check so we never anchor the split to a dead pane.
-caller_pane="${TMUX_PANE:-}"
-if [[ -z "$caller_pane" && -n "$lead_pane_file" && -r "$lead_pane_file" ]]; then
-  cand="$(head -n1 "$lead_pane_file" 2>/dev/null || true)"
-  if [[ -n "$cand" ]] && tmux display-message -p -t "$cand" '#{pane_id}' >/dev/null 2>&1; then
-    caller_pane="$cand"
-  fi
-fi
-if [[ -z "$caller_pane" ]]; then
-  caller_pane=$(tmux display-message -p '#{pane_id}' 2>/dev/null || true)
+# Resolve the pane THIS wrapper actually runs in by walking our ancestor PIDs
+# and matching tmux pane_pids (see lib/okstra/tmux-pane.sh). Reliable
+# regardless of $TMUX/$TMUX_PANE (stripped by Claude Code's Bash tool) and of
+# which tmux client is currently active — a bare `tmux display-message` would
+# instead return the most-recently-active client's pane, frequently a DIFFERENT
+# session than the okstra run, which is why earlier approaches mis-placed or
+# dropped the trace pane. Empty = not inside a tmux pane (e.g. Claude launched
+# from the GUI app) → the trace split below is skipped.
+caller_pane=""
+if type okstra_resolve_caller_pane >/dev/null 2>&1; then
+  caller_pane="$(okstra_resolve_caller_pane)"
 fi
 
 # Pane titles: worker (caller) pane gets `codex-<role>-<pid>`; the sibling

package/runtime/bin/okstra-gemini-exec.sh

@@ -132,28 +132,25 @@ status_path="${prompt_path%.md}.status.json"
 [[ "$status_path" == "$prompt_path" ]] && status_path="${prompt_path}.status.json"
 started_ts=$(date +%s)
 script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd -P)"
+# Trace-pane caller resolution helper (okstra_resolve_caller_pane). The lib dir
+# is a bin-sibling in both repo (scripts/lib/...) and installed
+# (~/.okstra/bin/lib/...) layouts; degrade silently if absent.
+[ -r "$script_dir/lib/okstra/tmux-pane.sh" ] && . "$script_dir/lib/okstra/tmux-pane.sh"
 python3 "$script_dir/okstra-wrapper-status.py" \
   init "$status_path" "$(basename "$0")" "$role" "$$" "$started_ts" "$log_path" \
   >>"$log_path" 2>&1 || true
 
 # Resolve the run dir and the trace-split anchor pane. See
-# `okstra-codex-exec.sh` for the full rationale — kept in lock-step: derive
-# `<RUN_DIR>` from the prompt path (paths.py SSOT) to read the lead-recorded
-# pane and to tag the trace pane; prefer that lead pane over the unreliable
-# active-pane probe (this wrapper runs backgrounded and `$TMUX`/`$TMUX_PANE`
-# are stripped).
+# `okstra-codex-exec.sh` / `lib/okstra/tmux-pane.sh` for the full rationale —
+# kept in lock-step: derive `<RUN_DIR>` from the prompt path (paths.py SSOT) to
+# tag the trace pane, and resolve the caller pane by walking our ancestor PIDs
+# against tmux pane_pids (reliable even though `$TMUX`/`$TMUX_PANE` are stripped
+# and the wrapper runs backgrounded). Empty = not inside a tmux pane → skip.
 run_dir="$(cd "$(dirname "$prompt_path")/.." 2>/dev/null && pwd -P || true)"
-lead_pane_file="${run_dir:+$run_dir/state/lead-pane.id}"
 
-caller_pane="${TMUX_PANE:-}"
-if [[ -z "$caller_pane" && -n "$lead_pane_file" && -r "$lead_pane_file" ]]; then
-  cand="$(head -n1 "$lead_pane_file" 2>/dev/null || true)"
-  if [[ -n "$cand" ]] && tmux display-message -p -t "$cand" '#{pane_id}' >/dev/null 2>&1; then
-    caller_pane="$cand"
-  fi
-fi
-if [[ -z "$caller_pane" ]]; then
-  caller_pane=$(tmux display-message -p '#{pane_id}' 2>/dev/null || true)
+caller_pane=""
+if type okstra_resolve_caller_pane >/dev/null 2>&1; then
+  caller_pane="$(okstra_resolve_caller_pane)"
 fi
 
 # Pane titles: worker (caller) pane gets `gemini-<role>-<pid>`; the sibling

package/runtime/bin/okstra-trace-cleanup.sh

@@ -37,6 +37,14 @@
 
 set -u
 
+# Trace-pane caller resolution helper (okstra_resolve_caller_pane) — see
+# lib/okstra/tmux-pane.sh. Used as the lead-pane fallback below so a missing /
+# stale lead-pane.id resolves to the pane THIS process actually runs in (via
+# ancestor-PID ↔ tmux pane_pid matching), never a foreign active-client pane.
+# Bin-sibling path in repo + installed layouts; degrade silently if absent.
+_clean_script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd -P)"
+[ -r "$_clean_script_dir/lib/okstra/tmux-pane.sh" ] && . "$_clean_script_dir/lib/okstra/tmux-pane.sh"
+
 MODE="kill"   # kill | list
 REAP=0
 run_dir=""
@@ -88,7 +96,11 @@ if [[ "$REAP" -eq 0 ]]; then
   [[ -r "$lead_pane_file" ]] && lead_pane="$(head -n1 "$lead_pane_file" 2>/dev/null || true)"
 fi
 if [[ -z "$lead_pane" ]] || ! tmux display-message -p -t "$lead_pane" '#{pane_id}' >/dev/null 2>&1; then
-  lead_pane="$(tmux display-message -p '#{pane_id}' 2>/dev/null || true)"
+  if type okstra_resolve_caller_pane >/dev/null 2>&1; then
+    lead_pane="$(okstra_resolve_caller_pane 2>/dev/null || true)"
+  else
+    lead_pane="$(tmux display-message -p '#{pane_id}' 2>/dev/null || true)"
+  fi
 fi
 
 # Does a trace pane's tag belong to the set we are closing?

package/runtime/prompts/profiles/_common-contract.md

@@ -30,8 +30,8 @@ profile document.
 - Anti-escalation rule (shared):
   - treating "다음 단계 진행해" or equivalent user phrases as authorisation to start a *different* lifecycle phase is forbidden. The next phase begins only in a separate okstra run launched with the new `--task-type`. Per-profile documents may further restrict this within their own scope.
 - Run-start pane recording (shared — runs ONCE at run start, before the FIRST worker dispatch):
-  - The wrappers anchor each trace pane to the lead's pane and the cleanup scopes the worker-agent scan to it, but Claude Code's Bash tool strips `$TMUX`/`$TMUX_PANE`, so the lead MUST record its own pane explicitly. Because the lead runs this in its OWN foreground pane, the active pane IS the lead's — reliable, unlike a backgrounded wrapper's later probe.
-  - The lead MUST run once, at run start: `mkdir -p "<RUN_DIR>/state" && tmux display-message -p '#{pane_id}' > "<RUN_DIR>/state/lead-pane.id" 2>/dev/null || true` (substitute the run's absolute `RUN_DIR`). Outside tmux this writes nothing and every pane step below silently no-ops — that empty/absent file is the single signal that the lead is not in tmux.
+  - The codex/gemini wrappers now self-anchor their trace pane by walking their own ancestor PIDs against tmux `pane_pid`s (see `lib/okstra/tmux-pane.sh`), so they no longer depend on this file. The lead still records its own pane id here for the cleanup steps below (which-pane-to-never-kill) and as the "am I in tmux" gate. A bare `tmux display-message -p '#{pane_id}'` is NOT reliable for this — Claude Code's Bash tool strips `$TMUX`/`$TMUX_PANE`, so that command returns the most-recently-active *client's* pane (often a different session, or a foreign pane when the lead is launched outside tmux entirely). The lead therefore records via the same ancestry resolver.
+  - The lead MUST run once, at run start: `mkdir -p "<RUN_DIR>/state" && { . "$HOME/.okstra/bin/lib/okstra/tmux-pane.sh" 2>/dev/null && okstra_resolve_caller_pane; } > "<RUN_DIR>/state/lead-pane.id" 2>/dev/null || true` (substitute the run's absolute `RUN_DIR`). When the lead is not inside a tmux pane (e.g. Claude launched from the GUI app) no ancestor matches a pane, the file is empty, and every pane step below silently no-ops — that empty/absent file is the single signal that the lead is not in tmux.
 - Phase-start pane reset (shared — runs BEFORE dispatching each new worker batch):
   - okstra creates two kinds of tmux pane per run: (a) **worker-agent panes** the harness gives to dispatched subagents (titled `claude-worker` / `codex-worker` / `gemini-worker` / `report-writer-worker`), and (b) **trace panes** the codex/gemini wrappers spawn (`<cli>-<role>-<pid>-tail`). Both accumulate across internal phases because each new phase dispatches a fresh worker batch and the prior panes are never reclaimed.
   - When `<RUN_DIR>/state/lead-pane.id` is non-empty (the lead is in tmux), the lead MUST run `$HOME/.okstra/bin/okstra-trace-cleanup.sh --run-dir "<RUN_DIR>"` **immediately before** dispatching the next phase's workers — i.e. just before emitting each `PROGRESS: phase-5.5-convergence round=<N>` marker and just before `PROGRESS: phase-6-synthesis dispatching report-writer-worker`. This closes every prior-phase okstra pane (worker-agent + trace) for this run, while NEVER killing the lead's own pane.
@@ -41,8 +41,8 @@ profile document.
   - This step is **automatic and silent** — NO user prompt (workers are idle sessions that have already delivered their results; there is nothing for the user to preserve). It runs only when team-state's `teamCreate.status == "ok"` (Teams mode was actually used); in the no-`team_name` fallback there is no team to delete, so silent-skip.
   - Sequence (token-usage collection MUST already be complete — `TeamDelete` removes `~/.claude/teams/<team>/` + `~/.claude/tasks/<team>/` but NOT the `~/.claude/projects/` jsonls Phase 7 reads, yet the read MUST precede teardown):
     1. Read `~/.claude/teams/okstra-<task-key>/config.json` and, for every `members` entry whose name is not the lead, `SendMessage(to: <name>, message: { type: "shutdown_request" })` to terminate it gracefully.
-    2. Wait for the shutdown confirmations / idle notifications from all addressed teammates.
-    3. Call `TeamDelete()`. If it errors with an active-members message, a teammate has not finished shutting down — wait briefly and retry `TeamDelete()` once.
+    2. These workers already delivered their results and terminated when their `Agent()` dispatch returned (the lead's completion evidence is the returned output + the existing result/final-report file, not a teardown ack) — a terminated session emits NO shutdown confirmation. Treat `shutdown_request` as best-effort (fire-and-forget); the lead MUST NOT block waiting for acks from addressed teammates. Proceed immediately to step 3.
+    3. Call `TeamDelete()` — the single synchronization point for teardown. If it errors with an active-members message, one teammate is genuinely still shutting down: wait briefly, retry `TeamDelete()` once, then proceed regardless of the result. NEVER loop or re-send `shutdown_request`; teardown must never block run completion once the work and final report already exist.
   - Report it in one short line (e.g. `worker 6명 종료 + 팀 해제`) and proceed. Emit `PROGRESS: phase-7-teardown disbanding team` immediately before step 1.
 - Phase wrap-up — okstra pane disposition (shared, MUST be the *last* step before returning control to the user):
   - At run end the only residual okstra panes are the LAST phase's (e.g. the `report-writer-worker` agent pane and any codex/gemini trace pane). `okstra-trace-cleanup.sh --list --run-dir "<RUN_DIR>"` returns one tab-separated `<pane_id>\t<pane_title>` line per residual okstra pane (worker-agent + trace) for this run.

package/runtime/python/okstra_ctl/report_views.py

@@ -407,21 +407,12 @@ class _GroupedSpec:
     user_input_col: int = -1
 
 
-_FOLLOWUP_WIDE_PREFIXES: tuple[str, ...] = ("title", "scope", "reason")
-
-
 def _grouped_table_spec(
     header_cells: list[str], section_path: list[str]
 ) -> Optional[_GroupedSpec]:
-    """Return a ``_GroupedSpec`` for the three wide final-report tables
-    that benefit from the compact layout — Execution Status, §5
-    Clarification Items, §7 Follow-up Tasks — or ``None`` for every other
-    table (which keeps the default per-cell ``td-narrow`` rendering).
-
-    Each table is identified by stable header tokens (the i18n token/cost
-    columns are never used as anchors). ``wide_cols`` lists the long-prose
-    columns that must keep a guaranteed min-width; everything else short
-    collapses into the leading metadata cell."""
+    """Only §5 Clarification Items is grouped in the HTML view (it keeps the
+    interactive form and stays flat in the .md). All other narrative tables are
+    already rendered compactly by the template, so no grouping is applied here."""
     norm = [h.strip() for h in header_cells]
 
     def _spec(headline: int, wide: tuple[int, ...], **kw) -> _GroupedSpec:
@@ -429,12 +420,8 @@ def _grouped_table_spec(
         group = tuple(c for c in range(len(norm)) if c != headline and c not in wide_set)
         return _GroupedSpec(headline_col=headline, group_cols=group, wide_cols=wide, **kw)
 
-    # Execution Status by Agent — Agent … Summary of Key Findings.
-    if len(norm) >= 3 and norm[0] == "Agent" and norm[-1] == "Summary of Key Findings":
-        return _spec(0, (len(norm) - 1,), kind="plain")
-
-    # §5 Clarification Items — keep the interactive form, but collapse the
-    # short ID/Kind/Status/… columns and widen Statement + User input.
+    # §5 Clarification Items — keep the interactive form, and widen the three
+    # long-prose columns (Expected form is prose too, not a code column).
     if (
         any("Clarification Items" in h for h in section_path)
         and not _section_forbids_form(section_path)
@@ -444,9 +431,15 @@ def _grouped_table_spec(
     ):
         statement_col = next(i for i, h in enumerate(norm) if h.startswith("Statement"))
         user_input_col = norm.index("User input")
+        expected_col = next(
+            (i for i, h in enumerate(norm) if h.startswith("Expected form")), -1
+        )
+        wide_cols = tuple(
+            c for c in (expected_col, statement_col, user_input_col) if c >= 0
+        )
         return _spec(
             norm.index("ID"),
-            (statement_col, user_input_col),
+            wide_cols,
             kind="clarification",
             id_col=norm.index("ID"),
             kind_col=norm.index("Kind") if "Kind" in norm else -1,
@@ -455,16 +448,6 @@ def _grouped_table_spec(
             user_input_col=user_input_col,
         )
 
-    # §7 Follow-up Tasks — widen Title / Scope / Reason, collapse the rest.
-    if any("Follow-up Tasks" in h for h in section_path) and "ID" in norm:
-        wide = tuple(
-            i
-            for i, h in enumerate(norm)
-            if any(h.lower().startswith(p) for p in _FOLLOWUP_WIDE_PREFIXES)
-        )
-        if wide:
-            return _spec(norm.index("ID"), wide, kind="plain")
-
     return None
 
 
@@ -768,6 +751,10 @@ def _inline(text: str) -> str:
     out = _LINK_PATTERN.sub(
         lambda m: f'<a href="{m.group(2)}">{m.group(1)}</a>', out
     )
+    # Preserve explicit <br> line breaks used inside compact meta cells (the
+    # markdown source intentionally stacks short fields with <br>). html.escape
+    # above turned them into &lt;br&gt;; restore the tag.
+    out = out.replace("&lt;br&gt;", "<br>").replace("&lt;br/&gt;", "<br>").replace("&lt;br /&gt;", "<br>")
     return out
 
 

package/runtime/templates/project-docs/task-index.template.md

@@ -68,11 +68,4 @@ taskType: "{{FM_TASK_TYPE}}"
 
 ## Notes
 
-- 이 문서는 사람용 quick summary입니다.
-- canonical metadata는 항상 `task-manifest.json`을 기준으로 확인합니다.
-- 동일한 버그나 작업을 다시 이어갈 때는 같은 `Task Group`과 `Task ID`를 재사용합니다.
-- `requirements-discovery`는 work category와 다음 phase routing 결정을 남기기 위한 초기 triage 단계입니다.
-- run directory는 `runs/<task-type>/` 규칙을 사용하여 서로 다른 task-type 실행을 경로 수준에서 분리합니다.
-- run directory 내부는 `manifests/`, `state/`, `prompts/`, `reports/`, `status/`, `sessions/`, `worker-results/`처럼 유형별 하위 폴더로 나뉩니다.
-- current run의 `prompts/` 디렉터리는 lead prompt snapshot과 worker prompt history file의 canonical 저장 위치입니다.
-- run-level artifact와 result 파일명은 같은 task-type 재실행을 구분하기 위해 `-YYYY-MM-DD_HH-MM-SS` suffix를 사용합니다.
+- 이 문서는 사람이 빠르게 훑어보기 위한 요약입니다. 정본·상세 metadata 는 `task-manifest.json` 을 참조하세요.

package/runtime/templates/reports/final-report.template.md

@@ -53,10 +53,10 @@ approved: {{ frontmatter.approved | yaml_scalar }}
 
 {{ t("sectionIntro.ticketCoverage") }}
 
-| ID | Ticket ID | {{ t("columns.summary") }} | {{ t("columns.source") }} |
-|----|-----------|------------|----------------------------|
+| {{ t("columns.recordMeta") }} | {{ t("columns.summary") }} |
+|--------|------------|
 {% for row in summary -%}
-| {{ row.id }} | `{{ row.ticketId }}` | {{ row.summary }} | {{ row.source }} |
+| **{{ row.id }}**<br>Ticket: `{{ row.ticketId }}`<br>{{ t("columns.source") }}: {{ row.source }} | {{ row.summary }} |
 {% endfor %}
 
 {% if ticketCoverage.omit %}
@@ -76,17 +76,15 @@ approved: {{ frontmatter.approved | yaml_scalar }}
 {% endfor %}
 {%- endif %}
 
-{{ t("ticketCoverage.ruleNote") }}
-
 {% endif %}
 ## Execution Status by Agent
 
 {{ t("sectionIntro.executionStatus") }}
 
-| Agent | Role | Model | Status | {{ t("columns.rawTokens") }} | {{ t("columns.billableTokens") }} | {{ t("columns.cost") }} | Duration | Summary of Key Findings |
-|-------|------|-------|--------|-----------|-----------|------------|----------|-------------------------|
+| {{ t("columns.recordMeta") }} | Summary of Key Findings |
+|--------|-------------------------|
 {% for row in executionStatus -%}
-| {{ row.agent }} | {{ row.role }} | {{ row.model }} | {{ row.status }} | {{ row.totalTokens | format_int }}{% if row.cliTotalTokens %} (CLI: {{ row.cliTotalTokens | format_int }}){% endif %} | {{ row.billableTokens | format_int }} | {{ row.costUsd | format_usd }}{% if row.cliCostUsd %} (+ CLI {{ row.cliCostUsd | format_usd }}){% endif %} | {{ row.durationMs | format_duration_ms }} | {{ row.summary }} |
+| **{{ row.agent }}**<br>Role: {{ row.role }}<br>Model: {{ row.model }}<br>Status: {{ row.status }}<br>{{ t("columns.rawTokens") }}: {{ row.totalTokens | format_int }}{% if row.cliTotalTokens %} (CLI: {{ row.cliTotalTokens | format_int }}){% endif %}<br>{{ t("columns.billableTokens") }}: {{ row.billableTokens | format_int }}<br>{{ t("columns.cost") }}: {{ row.costUsd | format_usd }}{% if row.cliCostUsd %} (+ CLI {{ row.cliCostUsd | format_usd }}){% endif %}<br>Duration: {{ row.durationMs | format_duration_ms }} | {{ row.summary }} |
 {% endfor %}
 
 ## {{ t("tokenSummary.heading") }}
@@ -130,10 +128,10 @@ approved: {{ frontmatter.approved | yaml_scalar }}
 {% if crossVerification.consensus | length == 0 -%}
 {{ t("emptyState.consensusItems") }}
 {%- else %}
-| ID | Ticket ID | Statement | Source items (worker:item) | Evidence (path:line / log / worker report) |
-|----|-----------|-----------|----------------------------|---------------------------------------------|
+| {{ t("columns.recordMeta") }} | Statement | Evidence (path:line / log / worker report) |
+|--------|-----------|---------------------------------------------|
 {% for row in crossVerification.consensus -%}
-| {{ row.id }} | `{{ row.ticketId }}` | {{ row.statement }} | {{ row.sourceItems | join(', ') }} | {{ row.evidence }} |
+| **{{ row.id }}**<br>Ticket: `{{ row.ticketId }}`<br>Source items: {{ row.sourceItems | join(', ') }} | {{ row.statement }} | {{ row.evidence }} |
 {% endfor %}
 {%- endif %}
 
@@ -144,10 +142,10 @@ approved: {{ frontmatter.approved | yaml_scalar }}
 {% if crossVerification.differences | length == 0 -%}
 {{ t("emptyState.differences") }}
 {%- else %}
-| ID | Ticket ID | Disagreement | Workers (position + item) | Evidence |
-|----|-----------|--------------|---------------------------|----------|
+| {{ t("columns.recordMeta") }} | Disagreement | Evidence |
+|--------|--------------|----------|
 {% for row in crossVerification.differences -%}
-| {{ row.id }} | `{{ row.ticketId }}` | {{ row.disagreement }} | {% for w in row.workersPosition %}{{ w.worker }}:{{ w.itemId }} ({{ w.position }}){% if not loop.last %} / {% endif %}{% endfor %} | {{ row.evidence }} |
+| **{{ row.id }}**<br>Ticket: `{{ row.ticketId }}`<br>Workers: {% for w in row.workersPosition %}{{ w.worker }}:{{ w.itemId }} ({{ w.position }}){% if not loop.last %} / {% endif %}{% endfor %} | {{ row.disagreement }} | {{ row.evidence }} |
 {% endfor %}
 {%- endif %}
 
@@ -170,10 +168,10 @@ approved: {{ frontmatter.approved | yaml_scalar }}
 {% if evidence.primary | length == 0 -%}
 {{ t("emptyState.primaryEvidence") }}
 {%- else %}
-| ID | Ticket ID | Evidence | Source items (worker:item) | Source (path:line / log) |
-|----|-----------|----------|----------------------------|---------------------------|
+| {{ t("columns.recordMeta") }} | Evidence |
+|--------|----------|
 {% for row in evidence.primary -%}
-| {{ row.id }} | `{{ row.ticketId }}` | {{ row.evidence }} | {{ row.sourceItems | join(', ') }} | {{ row.source }} |
+| **{{ row.id }}**<br>Ticket: `{{ row.ticketId }}`<br>Source items: {{ row.sourceItems | join(', ') }}<br>Source: {{ row.source }} | {{ row.evidence }} |
 {% endfor %}
 {%- endif %}
 
@@ -184,10 +182,10 @@ approved: {{ frontmatter.approved | yaml_scalar }}
 {% if not evidence.secondary or evidence.secondary | length == 0 -%}
 {{ t("emptyState.secondaryEvidence") }}
 {%- else %}
-| ID | Ticket ID | Hypothesis or supporting evidence | Source / confidence |
-|----|-----------|-----------------------------------|---------------------|
+| {{ t("columns.recordMeta") }} | Hypothesis or supporting evidence | Source / confidence |
+|--------|-----------------------------------|---------------------|
 {% for row in evidence.secondary -%}
-| {{ row.id }} | `{{ row.ticketId }}` | {{ row.hypothesis }} | {{ row.confidence }} |
+| **{{ row.id }}**<br>Ticket: `{{ row.ticketId }}` | {{ row.hypothesis }} | {{ row.confidence }} |
 {% endfor %}
 {%- endif %}
 
@@ -196,10 +194,10 @@ approved: {{ frontmatter.approved | yaml_scalar }}
 {% if missingInformation | length == 0 -%}
 {{ t("emptyState.risks") }}
 {%- else %}
-| ID | Ticket ID | Item | Risk if ignored | Mitigation Owner |
-|----|-----------|------|-----------------|------------------|
+| {{ t("columns.recordMeta") }} | Item | Risk if ignored | Mitigation Owner |
+|--------|------|-----------------|------------------|
 {% for row in missingInformation -%}
-| {{ row.id }} | `{{ row.ticketId }}` | {{ row.item }} | {{ row.risk }} | {{ row.owner }} |
+| **{{ row.id }}**<br>Ticket: `{{ row.ticketId }}` | {{ row.item }} | {{ row.risk }} | {{ row.owner }} |
 {% endfor %}
 {%- endif %}
 
@@ -249,8 +247,6 @@ approved: {{ frontmatter.approved | yaml_scalar }}
 | {{ row.step }} | `{{ row.ticketId }}` | {{ row.action }} | `{{ row.files }}` | `{{ row.commandOrTest }}` | {{ row.expectedOutcome }} |
 {% endfor %}
 
-{{ t("sectionIntro.stepRule") }}
-
 ### 4.5.5 Dependency / Migration Risk{% if t("sectionAside.dependencyRisk") != "Dependency / Migration Risk" %} ({{ t("sectionAside.dependencyRisk") }}){% endif %}
 
 {% if implementationPlanning.dependencyMigrationRisk | length == 0 -%}
@@ -598,9 +594,9 @@ approved: {{ frontmatter.approved | yaml_scalar }}
 {% if followUpTasks | length == 0 -%}
 {{ t("emptyState.noFollowUp") }}
 {%- else %}
-| ID | Ticket ID | Origin | New Task ID | Title | Suggested task-type | Scope (files/areas) | Reason / Why deferred | Priority (P0/P1/P2) | Auto-spawn? |
-|----|-----------|--------|-------------|-------|---------------------|---------------------|------------------------|---------------------|-------------|
+| {{ t("columns.recordMeta") }} | Title | Scope (files/areas) | Reason / Why deferred |
+|--------|-------|---------------------|------------------------|
 {% for row in followUpTasks -%}
-| {{ row.id }} | `{{ row.ticketId }}` | `{{ row.origin }}` | `{{ row.newTaskId }}` | {{ row.title }} | `{{ row.suggestedTaskType }}` | `{{ row.scope }}` | {{ row.reason }} | `{{ row.priority }}` | `{{ row.autoSpawn }}` |
+| **{{ row.id }}**<br>Ticket: `{{ row.ticketId }}`<br>Origin: `{{ row.origin }}`<br>New Task ID: `{{ row.newTaskId }}`<br>Type: `{{ row.suggestedTaskType }}`<br>Priority: `{{ row.priority }}`<br>Auto-spawn: `{{ row.autoSpawn }}` | {{ row.title }} | {{ row.scope }} | {{ row.reason }} |
 {% endfor %}
 {%- endif %}

package/runtime/templates/reports/i18n/en.json

@@ -19,6 +19,7 @@
     "noFollowUp": "- No follow-up tasks. The next phase for this run is in §6 (Recommended Next Steps)."
   },
   "columns": {
+    "recordMeta": "Record",
     "summary": "Summary",
     "source": "Source (brief/source/worker)",
     "rawTokens": "Raw tokens",
@@ -38,14 +39,13 @@
     "stepwiseExecutionOrder": "Stepwise Execution Order"
   },
   "sectionIntro": {
-    "verdictCard": "At-a-glance verdict card. Every value in this table MUST exactly match the authoritative values in `## 2. Final Verdict` and `## 6. Recommended Next Steps`.",
-    "clarificationCarryIn": "Walk every C-* row of the prior report's `## 5. Clarification Items` table against new evidence. Update each row's `Status` to `resolved` or `obsolete` and carry it into this run's `## 5.` table. Cite the resolution evidence (file:line / log / worker result) inline.",
-    "ticketCoverage": "Summarize 3-5 core problems / requirements / verification targets as a table. Base this on the brief, source material, and worker results.",
-    "executionStatus": "Aggregate each worker's status, assigned model, and key findings into one table. Do NOT replace worker artifacts with ungrounded claims.",
-    "sourceItemsRule": "`Source items` rule: list which worker items this consensus row was synthesised from, as a comma-list of `<worker>:<item-id>` pairs. Full policy: `prompts/profiles/_common-contract.md` \"Cross-worker traceability\" SSOT.",
-    "stepRule": "Rule: each step is roughly 2-5 minutes. Every step MUST include exact file paths and commands.",
-    "planBodyVerification": "Result of Phase 6 — report-writer's synthesised 4.5 body re-cast to workers by the lead on a plan-item basis, with verdicts collected.",
-    "clarificationItems": "Track items that must be answered by the user or backed by attached material before the next run advances, **all inside one table**."
+    "verdictCard": "At-a-glance verdict card — a summary that mirrors the values in `## 2. Final Verdict` and `## 6. Recommended Next Steps`.",
+    "clarificationCarryIn": "Unresolved `Clarification Items` from the prior report, re-examined against new evidence and carried in with updated status.",
+    "ticketCoverage": "Summary of the core problems, requirements, and verification targets this run covered.",
+    "executionStatus": "At-a-glance table of each worker's status, assigned model, and key findings.",
+    "sourceItemsRule": "The `Source items` column shows which worker items each consensus row was synthesised from, as `<worker>:<item-id>` pairs.",
+    "planBodyVerification": "Cross-verification results for each item in the plan body.",
+    "clarificationItems": "Items that need your answer or supporting material before the next step."
   },
   "tokenSummary": {
     "heading": "Token Usage Summary",
@@ -63,19 +63,18 @@
     "nextStepLabel": "Next step"
   },
   "ticketCoverage": {
-    "intro": "Reverse index of tickets covered in this run. All body items are linked to a ticket via the `Ticket ID` column or `[TICKETID: <id>]` tag.",
+    "intro": "The tickets this run covered, with the sections and related items where each ticket appears.",
     "columnSections": "Sections",
-    "columnRelatedIds": "Related item IDs",
-    "ruleNote": "Rule: `Ticket ID` must match the ticket key exactly as it appears in the body. When `Issue / Ticket` is empty and falls back, use the `Task ID` value as-is without prefix (e.g. `8852`). Use `unknown` when not identifiable."
+    "columnRelatedIds": "Related item IDs"
   },
   "finalVerdict": {
-    "intro": "State the final conclusion and recommended direction in one table. `Direction` ∈ `continue-investigation / begin-implementation / approve / reject / hold`. When `task-type` is `final-verification`, `Verdict Token` must be one of `accepted / conditional-accept / blocked`; `release-handoff` uses this value as an entry gate. For all other task-types use `not-applicable`."
+    "intro": "Final conclusion and recommended direction. `Direction` values: `continue-investigation / begin-implementation / approve / reject / hold`. When `task-type` is `final-verification`, `Verdict Token` is one of `accepted / conditional-accept / blocked` and serves as the `release-handoff` entry gate. For all other task-types: `not-applicable`."
   },
   "evidence": {
-    "sourceItemsColumnNote": "`Source items` column rule is the same as §1.1."
+    "sourceItemsColumnNote": "The `Source items` column is described in §1.1."
   },
   "roundHistory": {
-    "round2SkippedReasonNote": "value is one of `queue-empty | max-rounds-1 | all-reverify-non-result | not-skipped | convergence-disabled | single-analyser-only`",
+    "round2SkippedReasonNote": "(one of: `queue-empty`, `max-rounds-1`, `all-reverify-non-result`, `not-skipped`, `convergence-disabled`, `single-analyser-only`)",
     "singleRoundPrefix": "Single round —",
     "noRoundsNote": "No reverify rounds executed (all findings reached consensus at grouping)."
   },
@@ -128,7 +127,7 @@
   "clarification": {
     "fillAndRerun": "Fill in your answers then re-run the same phase:",
     "separateTerminalLabel": "Separate terminal",
-    "columnGuide": "Column guide (full definitions: `prompts/profiles/_common-contract.md §Clarification request policy` SSOT):"
+    "columnGuide": "Column descriptions:"
   },
   "followUpTasks": {
     "headingAside": "Follow-up Tasks"

package/runtime/templates/reports/i18n/ko.json

@@ -19,6 +19,7 @@
     "noFollowUp": "- 후속 작업 없음. 본 run 의 다음 phase 는 §6 (Recommended Next Steps) 참고."
   },
   "columns": {
+    "recordMeta": "항목",
     "summary": "한 줄 요약",
     "source": "출처 (brief/source/worker)",
     "rawTokens": "처리 토큰",
@@ -38,14 +39,13 @@
     "stepwiseExecutionOrder": "단계별 실행 순서"
   },
   "sectionIntro": {
-    "verdictCard": "한눈에 보는 결과 카드. 본 표의 모든 값은 `## 2. Final Verdict` 및 `## 6. Recommended Next Steps` 의 권위 있는 값과 정확히 일치해야 합니다.",
-    "clarificationCarryIn": "이전 보고서의 `## 5. Clarification Items` 표 매 행(`C-001`, `C-002`, …) 을 새 증거에 비추어 검토하고, 각 행의 `Status` 를 `resolved` 또는 `obsolete` 로 갱신한 뒤 본 run 의 `## 5.` 표에 carry-in 합니다. 해소 근거(파일:라인 / 로그 / 워커 결과) 를 함께 인용합니다.",
-    "ticketCoverage": "3~5 개 row 로 핵심 문제·요구사항·검증 대상을 표로 정리합니다. brief, 소스 자료, worker 결과를 근거로 작성합니다.",
-    "executionStatus": "각 worker 의 status, 배정 모델, key finding 을 한 표에 모읍니다. worker 산출물을 근거 없는 주장으로 대체하지 않습니다.",
-    "sourceItemsRule": "`Source items` 규칙: 본 합의 row 가 어느 워커의 어느 항목들에서 합성됐는지를 `<worker>:<item-id>` 페어 콤마-리스트로 적습니다. 자세한 정책은 `prompts/profiles/_common-contract.md` \"Cross-worker traceability\" SSOT.",
-    "stepRule": "규칙: 한 step 은 약 2~5 분. 모든 step 은 정확한 파일 경로와 명령어 포함.",
-    "planBodyVerification": "Phase 6 에서 report-writer 가 합성한 4.5 본문을 lead 가 plan-item 단위로 워커들에게 다시 던지고 평결을 수집한 결과.",
-    "clarificationItems": "다음 run 으로 넘어가기 전에 사용자가 답하거나 자료를 첨부해야 하는 항목을 **한 표 안에서** 추적합니다."
+    "verdictCard": "한눈에 보는 결과 카드 — `## 2. Final Verdict` 와 `## 6. Recommended Next Steps` 의 값을 그대로 옮긴 요약입니다.",
+    "clarificationCarryIn": "이전 보고서의 미해결 `Clarification Items` 를 새 증거로 재검토해 해소·갱신한 뒤 이어받은 항목입니다.",
+    "ticketCoverage": "이 run 이 다룬 핵심 문제·요구사항·검증 대상 요약입니다.",
+    "executionStatus": "각 worker 의 상태·배정 모델·핵심 finding 을 한눈에 보는 표입니다.",
+    "sourceItemsRule": "`Source items` 열은 이 합의 항목이 어느 워커의 어느 항목에서 합성됐는지를 `<worker>:<item-id>` 형식으로 표기합니다.",
+    "planBodyVerification": "계획 본문의 각 항목을 plan-item 단위로 교차 검증한 결과입니다.",
+    "clarificationItems": "다음 진행 전에 사용자의 답변이나 자료 첨부가 필요한 항목입니다."
   },
   "tokenSummary": {
     "heading": "토큰 사용량 요약",
@@ -63,19 +63,18 @@
     "nextStepLabel": "다음 단계"
   },
   "ticketCoverage": {
-    "intro": "본 run 이 다룬 ticket 의 역방향 인덱스. 본문 항목들은 모두 `Ticket ID` 컬럼 또는 `[TICKETID: <id>]` 태그로 ticket 과 묶여 있습니다.",
+    "intro": "이 run 이 다룬 ticket 과, 각 ticket 이 등장한 섹션·관련 항목 목록입니다.",
     "columnSections": "등장 섹션",
-    "columnRelatedIds": "관련 항목 IDs",
-    "ruleNote": "규칙: `Ticket ID` 는 본문에서 등장한 ticket 키와 정확히 동일 문자열. `Issue / Ticket` 이 비어 폴백된 경우 `Task ID` 값을 prefix 없이 그대로 (예: `8852`). 식별 불가는 `unknown`."
+    "columnRelatedIds": "관련 항목 IDs"
   },
   "finalVerdict": {
-    "intro": "최종 결론과 권장 방향을 한 표로 명시합니다. `Direction` ∈ `continue-investigation / begin-implementation / approve / reject / hold`. `task-type` 이 `final-verification` 이면 `Verdict Token` 은 `accepted / conditional-accept / blocked` 중 하나여야 하며, `release-handoff` 는 이 값을 진입 게이트로 사용합니다. 다른 task-type 에서는 `not-applicable`."
+    "intro": "최종 결론과 권장 방향입니다. `Direction` 값: `continue-investigation / begin-implementation / approve / reject / hold`. `task-type` 이 `final-verification` 일 때 `Verdict Token` 은 `accepted / conditional-accept / blocked` 중 하나이며 `release-handoff` 진입 게이트로 쓰입니다. 그 외 task-type 에서는 `not-applicable`."
   },
   "evidence": {
-    "sourceItemsColumnNote": "`Source items` 컬럼 규칙은 §1.1 과 동일."
+    "sourceItemsColumnNote": "`Source items` 열 설명은 §1.1 과 동일합니다."
   },
   "roundHistory": {
-    "round2SkippedReasonNote": "값은 `queue-empty | max-rounds-1 | all-reverify-non-result | not-skipped | convergence-disabled | single-analyser-only` 중 하나.",
+    "round2SkippedReasonNote": "(허용 값: `queue-empty`, `max-rounds-1`, `all-reverify-non-result`, `not-skipped`, `convergence-disabled`, `single-analyser-only`)",
     "singleRoundPrefix": "단일 라운드 —",
     "noRoundsNote": "재검증 라운드 미실행 (그룹핑 단계에서 전부 합의)."
   },
@@ -128,7 +127,7 @@
   "clarification": {
     "fillAndRerun": "답을 채우신 뒤 같은 phase 를 다시 실행:",
     "separateTerminalLabel": "별도 터미널",
-    "columnGuide": "컬럼 가이드 (전체 정의는 `prompts/profiles/_common-contract.md §Clarification request policy` SSOT 참조):"
+    "columnGuide": "각 컬럼 설명:"
   },
   "followUpTasks": {
     "headingAside": "후속 작업"

package/runtime/templates/reports/schedule.template.md

@@ -82,7 +82,8 @@ _의존 정보 없음_
      If included, MUST use this exact heading literal `## Gantt Chart`.
      Render as ASCII inside a fenced ``` block with NO language tag.
      Mermaid / PlantUML / Graphviz are forbidden — see SKILL §"ASCII Gantt format".
-     The axis is RELATIVE DAY-COUNTS (Day 1 / J1, …) — never calendar dates. -->
+     The axis is RELATIVE DAY-COUNTS (Day 1 / J1, …) — never calendar dates.
+     If day estimates are missing, omit this whole section and replace it with one blockquote: `> _Gantt Chart 생략: <reason>_` (SKILL §"MUST — emit a skip-reason note"). -->
 
 ```
 Day:            1    5    10   15   20   25   30
@@ -96,12 +97,7 @@ Phase 3
   <TASK-ID> (<size>)                                  ████  (after <TASK-ID>)
 ```
 
-> 가로축은 **상대 일수** (Day 1 시작 = Phase 1 시작). 오늘 날짜·요일·달력일자 사용 금지.
-> 임계 경로(critical path) 항목은 `! crit` 주석, 추정 배분은 `est` 주석으로 표기.
-> 막대는 `█`(확정) + `░`(상한/불확실) 조합. 단일 mid-point bar 도 허용.
-> 데이터가 없어 이 섹션을 생략할 때는, `## Gantt Chart` 헤딩과 ASCII 블록 전체를
-> 다음 한 줄 blockquote로 대체한다 (SKILL §"MUST — emit a skip-reason note" 참조):
-> `> _Gantt Chart 생략: <구체 사유 — 예: "단일 task (DEV-9187) + XL effort 5–15d 범위로 정밀 day 추정 부재."_`
+> 가로축은 **상대 일수**입니다 (Day 1 = Phase 1 시작). 범례: `! crit` = 임계 경로(critical path), `est` = 추정 배분, `█` = 확정 구간, `░` = 상한/불확실 구간.
 
 ---