@hunter-harness/workflow-harness 0.2.15 → 0.2.17

package/harness/bundles/general/claude-code/harness-sync/SKILL.md

@@ -19,7 +19,7 @@ disallowed-tools:
   - Bash(awk *)
   - Bash(curl *)
 ---
-<!-- generated by harness_deploy.py; core=8fbcf97a5a4408df; overlay=none; agent=claude-code; do not edit -->
+<!-- generated by harness_deploy.py; core=d5b51ad56d4d4535; overlay=none; agent=claude-code; do not edit -->
 # harness-sync — 元数据同步
 
 ## Purpose
@@ -55,7 +55,7 @@ python <skills-root>/scripts/harness_sync.py finalize --project . --run-id <run-
 | 3 | `.harness/codebase/map/` 是否过期 → 报告建议 `/harness-codebase-map` |
 | 3.5 | `harness_knowledge.py sync`；可 `sync --update`；失败不得假装可用。**知识闭环主入口**是 `/harness-knowledge-ingest auto`（含 Agent judge），sync 不重复列人工知识待办 |
 | 3.6 | 单次运行 `harness_knowledge.py maintain --project . --drain --json`，有界推进全部 `.harness/knowledge/maintenance-outbox/{pending,failed}`；不得为每个条目重复启动 Python/重建索引（§8.2：archive close 只 enqueue，sync 异步推进 outbox 到 completed/pending-judge） |
-| 4 | CLAUDE.md 完整性/行数 → 超限 AskUserQuestion 瘦身 |
+| 4 | CLAUDE.md 完整性/行数 → 超限 blocking user confirmation 瘦身 |
 | 5 | AGENTS.md 与各 Agent 指令入口一致；Claude Code 启用时验证 CLAUDE.md 引用 AGENTS.md，禁止反向循环引用 |
 | 6 | `.harness/` 结构（init 规程 → `reference.md` 第 6 步）；已装 skill 新鲜度只经 `hunter-harness refresh --dry-run --json` 的 post-adaptation freshness 判断，禁止 raw build 字节比较 |
 | 7–9 | `项目规则（见 .harness/context-index.json）/`、构建配置、测试目录 — **只提示不自动修复** |
@@ -67,6 +67,7 @@ python <skills-root>/scripts/harness_sync.py finalize --project . --run-id <run-
 - 命令结果不得靠猜测；普通 Bash 被拒 → 立即改用等价 PowerShell 重试一次
 - 仅 PowerShell 成功且有明确证据（构建/git/测试输出、文件存在、exit 0）时可标 ✅OK；否则 ❌FAIL 或 🟡WARN
 - 禁止把 hook 拒绝、静态验证、无输出、用户跳过说成成功 → 详见 [[../protocols/powershell-protocol.md|powershell-protocol]]、[[../protocols/evidence-based-reporting-protocol.md|evidence-based-reporting-protocol]]
+> 片段：p0-trust
 
 ## 关键规则
 
@@ -97,3 +98,4 @@ python <skills-root>/scripts/harness_events.py append --change-dir ".harness/cha
 > **Task 4 §6.1 写入契约**：普通 `append` = 加锁 -> 追加一行 -> fsync -> 解锁，**不 load 历史、不渲染**（O(1)，跨进程锁 `events.ndjson.lock`，UUID 用完整 `uuid4().hex` 无需去重扫描）。仅 `--type phase.end` append 在追加成功后渲染一次 `execution-log.md`；显式 `harness_events.py render` 随时从完整 events 重建；`harness_archive.py finalize` 在 collect 前强制 render 一次。高频 command append 期间 log 可能滞后，phase 边界保持最新。
 
 每次阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`；重试同一阶段时增加 `--attempt <n>`，不得覆盖或伪装成一次执行。跨工具继续执行时写 `--executor-tool <codex|claude-code|codebuddy|cursor>`，并在接棒事件写 `--handoff-from-tool` / `--handoff-reason`；也可由 `HUNTER_HARNESS_TOOL/AGENT/MODEL/RUN_ID` 环境变量统一注入。
+> 片段：logging · phase=`sync`；有未归档变更时写入其 change-dir events

package/harness/bundles/general/claude-code/harness-sync/reference.md

@@ -58,7 +58,7 @@ powershell.exe -Command "git -C '<项目路径>' diff --stat HEAD~5 2>$null"
 
 **瘦身策略**：如果超过 200 行，不是简单截断——而是把可独立成篇的细节（如完整编码规范、详细架构图）迁移到 `项目规则（见 .harness/context-index.json）/` 下，CLAUDE.md 保留简洁的索引和关键命令。
 
-> ⚠️ CLAUDE.md 需要瘦身时，必须用 AskUserQuestion 与用户确认拆分方案后再执行。
+> ⚠️ CLAUDE.md 需要瘦身时，必须用 blocking user confirmation 与用户确认拆分方案后再执行。
 
 ### 5. AGENTS.md 一致性
 

package/harness/bundles/general/claude-code/harness-test/SKILL.md

@@ -22,7 +22,7 @@ disallowed-tools:
   - Bash(node *)
   - Bash(codegraph *)
 ---
-<!-- generated by harness_deploy.py; core=8fbcf97a5a4408df; overlay=none; agent=claude-code; do not edit -->
+<!-- generated by harness_deploy.py; core=d5b51ad56d4d4535; overlay=none; agent=claude-code; do not edit -->
 # harness-test — 测试执行
 
 ## Purpose
@@ -59,6 +59,7 @@ disallowed-tools:
 9. **每个 skill 边界只刷新一次状态快照** — 先运行 `python <skills-root>/scripts/harness_state.py capture --project . --change-dir ".harness/changes/<change-name>" --base <baseCommit> --json`；`changedSegments=[]` 时复用已有 profile/rules/map/knowledge/change/code 指纹，不再重复跑全量 `harness-sync`、知识重建或代码库扫描。只有对应 segment 变化或缺失时才定向刷新；快照不能替代源码读取和验证门禁。
 
 状态目录分层：新路径优先，旧路径兼容 → [[../protocols/state-layout-protocol.md|state-layout-protocol]]
+> 片段：read-protocol
 
 ## Workflow
 
@@ -107,6 +108,7 @@ disallowed-tools:
 - 命令结果不得靠猜测；普通 Bash 被拒 → 立即改用等价 PowerShell 重试一次
 - 仅 PowerShell 成功且有明确证据（构建/git/测试输出、文件存在、exit 0）时可标 ✅OK；否则 ❌FAIL 或 🟡WARN
 - 禁止把 hook 拒绝、静态验证、无输出、用户跳过说成成功 → 详见 [[../protocols/powershell-protocol.md|powershell-protocol]]、[[../protocols/evidence-based-reporting-protocol.md|evidence-based-reporting-protocol]]
+> 片段：p0-trust
 
 ## 关键规则（硬门禁速查）
 
@@ -146,11 +148,11 @@ python <skills-root>/scripts/harness_test_guard.py record --project . --change-d
 
 ### 六、服务启动 + 生命周期管理
 
-启动等待状态机：0–30s 每 2s 探测、30–120s 每 5s 探测、>120s 读日志判定；遇启动失败特征立即停。**Service Gate**：`harness_service.py ensure` 返回 `action=needs-user-decision`（用户自启服务占端口）时 **才** AskUserQuestion；AI 托管服务或端口空闲则自动继续，不询问。服务指纹（`moduleInputsHash`，来自 CLI `--files` ∪ `serviceStart.inputFiles`）+ `startCommandHash` + `profile` + `overlayPath` + 进程身份任一变化即 restart；**空输入被拒绝**，不生成可复用空指纹。测试结束默认清理 AI 启动的服务。详见 `reference.md`「服务决策门」。
+启动等待状态机：0–30s 每 2s 探测、30–120s 每 5s 探测、>120s 读日志判定；遇启动失败特征立即停。**Service Gate**：`harness_service.py ensure` 返回 `action=needs-user-decision`（用户自启服务占端口）时 **才** blocking user confirmation；AI 托管服务或端口空闲则自动继续，不询问。服务指纹（`moduleInputsHash`，来自 CLI `--files` ∪ `serviceStart.inputFiles`）+ `startCommandHash` + `profile` + `overlayPath` + 进程身份任一变化即 restart；**空输入被拒绝**，不生成可复用空指纹。测试结束默认清理 AI 启动的服务。详见 `reference.md`「服务决策门」。
 
 ### 七、运行时配置叠加（不动 tracked 配置）
 
-禁止默认 Edit tracked 应用配置文件。默认运行时配置叠加（ASCII 绝对路径）；改 tracked 配置 → 默认拒绝，记 `decision` 事件（不 AskUserQuestion，报告 🟡 WARN）。详见 `reference.md`。
+禁止默认 Edit tracked 应用配置文件。默认运行时配置叠加（ASCII 绝对路径）；改 tracked 配置 → 默认拒绝，记 `decision` 事件（不 blocking user confirmation，报告 🟡 WARN）。详见 `reference.md`。
 
 ### 八、Token 缓存与复用
 
@@ -182,7 +184,7 @@ python <skills-root>/scripts/harness_test_guard.py record --project . --change-d
 
 ## 交互白名单
 
-本 skill **仅允许**以下 AskUserQuestion；其余默认值 + `decision` 事件：
+本 skill **仅允许**以下 blocking user confirmation；其余默认值 + `decision` 事件：
 
 1. **Service Gate**：仅当 `harness_service.py ensure` 返回 `needs-user-decision`（用户进程占端口）时询问处理方式
 
@@ -199,3 +201,4 @@ python <skills-root>/scripts/harness_events.py append --change-dir ".harness/cha
 > **Task 4 §6.1 写入契约**：普通 `append` = 加锁 -> 追加一行 -> fsync -> 解锁，**不 load 历史、不渲染**（O(1)，跨进程锁 `events.ndjson.lock`，UUID 用完整 `uuid4().hex` 无需去重扫描）。仅 `--type phase.end` append 在追加成功后渲染一次 `execution-log.md`；显式 `harness_events.py render` 随时从完整 events 重建；`harness_archive.py finalize` 在 collect 前强制 render 一次。高频 command append 期间 log 可能滞后，phase 边界保持最新。
 
 每次阶段结束必须写 `--status OK|WARN|FAIL|BLOCKED`；重试同一阶段时增加 `--attempt <n>`，不得覆盖或伪装成一次执行。跨工具继续执行时写 `--executor-tool <codex|claude-code|codebuddy|cursor>`，并在接棒事件写 `--handoff-from-tool` / `--handoff-reason`；也可由 `HUNTER_HARNESS_TOOL/AGENT/MODEL/RUN_ID` 环境变量统一注入。
+> 片段：logging · phase=`test` · 事件：phase/command/verification/decision/issue/artifact

package/harness/bundles/general/claude-code/harness-test/checklist.md

@@ -276,7 +276,7 @@ powershell.exe -NoProfile -Command "try { (Invoke-WebRequest -Uri 'http://127.0.
 - [ ] **只有首选执行器不可执行才降级**：PowerShell batch (.ps1) → 多次 MCP `browser_evaluate` → curl
 - [ ] **如果执行器在 PowerShell 可用，禁止使用 Playwright MCP 逐条执行接口测试**
 - [ ] 执行器直连本地 baseURL，复用认证凭证，不依赖浏览器 origin
-- [ ] 记录每个请求的 durationMs、status、code、message、assertionResult
+- [ ] 记录每个请求的 requestDurationMs、status、code、message、assertionResult
 - [ ] 支持失败时继续执行后续非依赖场景
 - [ ] 报告区分四种执行器：**接口测试执行器 / PowerShell batch / Playwright MCP browser_evaluate / curl**（不得与"Playwright"混写）
 
@@ -287,11 +287,14 @@ powershell.exe -NoProfile -Command "try { (Invoke-WebRequest -Uri 'http://127.0.
 - [ ] 测试结束后清理可清理的数据
 - [ ] 不能清理的记录遗留数据和原因
 
-### 请求耗时
+### 请求耗时（§5.22 分层耗时）
 
-- [ ] 每个请求记录 durationMs
-- [ ] durationMs > 10000 → 🟡SLOW，说明原因
-- [ ] durationMs > 30000 → ❌TIMEOUT_RISK，说明原因
+- [ ] 每个请求记录 requestDurationMs（单 HTTP 请求耗时）
+- [ ] 整个批次记录 batchDurationMs（runner wall-clock）
+- [ ] requestDurationMs > 10000 → 🟡SLOW，说明原因
+- [ ] requestDurationMs > 30000 → ❌TIMEOUT_RISK，说明原因
+- [ ] batchDurationMs > 180000 → 🟡BATCH_SLOW
+- [ ] 聚合合同套件场景只记录 batch reference + coveredTests，**禁止均摊生成伪 requestDurationMs**
 
 ### 覆盖标注诚实性（✅ 实测 vs 🟡 推断）
 

package/harness/bundles/general/claude-code/harness-test/pitfalls.md

@@ -191,7 +191,7 @@ $cred = $resp.data.accessToken
 **严重度**：🟡WARN
 **场景**：测试期间为了切换外部服务配置直接 Edit tracked 配置文件（如 `application-local-dev.yml`），测试后再还原，留下 diff 噪声
 **后果**：git diff --stat 出现意外文件、提交风险、误以为有业务变更
-**正确做法**：生成运行时配置叠加 `.harness/changes/<change>/runtime/application-harness-test.yml`（不提交），启动用 `-Dspring.config.additional-location=file:...` 叠加（按技术栈）；如必须改 tracked 配置，先 AskUserQuestion，最终报告至少 🟡 WARN
+**正确做法**：生成运行时配置叠加 `.harness/changes/<change>/runtime/application-harness-test.yml`（不提交），启动用 `-Dspring.config.additional-location=file:...` 叠加（按技术栈）；如必须改 tracked 配置，先 blocking user confirmation，最终报告至少 🟡 WARN
 
 ### 规则 22：唯一约束字段与本地预存数据冲突
 **严重度**：❌FAIL

package/harness/bundles/general/claude-code/harness-test/reference.md

@@ -150,7 +150,7 @@ spring:
 powershell.exe -NoProfile -Command "mvn spring-boot:run -pl <module> -Dspring-boot.run.profiles=<profile> -Dspring-boot.run.jvmArguments='-Dspring.config.additional-location=file:C:/temp/harness-test-overlay/<change-name>/application-harness-test.yml'"
 ```
 
-不得默认使用 `.harness/changes/<change>/runtime/application-harness-test.yml` 相对路径作为 JVM `additional-location`。如必须修改 tracked 配置文件，先 AskUserQuestion，最终报告至少 🟡 WARN。
+不得默认使用 `.harness/changes/<change>/runtime/application-harness-test.yml` 相对路径作为 JVM `additional-location`。如必须修改 tracked 配置文件，先 blocking user confirmation，最终报告至少 🟡 WARN。
 
 ## 服务决策门（Service Decision Gate）与服务生命周期管理
 
@@ -361,11 +361,13 @@ async function cleanup(token) {
 
 // ===== 主流程 =====
 const token = await resolveCredential('admin');
+const batchStart = Date.now();
 await setup(token);
 for (const scenario of scenarios) {
   await runOne(scenario, token);
 }
 const cleanupResult = await cleanup(token);
+const batchDurationMs = Date.now() - batchStart;
 
 const summary = {
   runner: 'api-test-executor',
@@ -377,7 +379,11 @@ const summary = {
   setupErrors: setupState.errors,
   cleanupResult,
   credentialRefreshCount,
-  startedAt: '<ISO>', finishedAt: '<ISO>', durationMs: 0
+  // §5.22: 分层耗时 schema — batchDurationMs 是整个 runner wall-clock；
+  // 每个场景的 requestDurationMs 是单 HTTP 请求耗时。聚合合同套件场景
+  // 只记录 batch reference/coveredTests，不允许均摊生成伪请求耗时。
+  batchDurationMs,
+  startedAt: '<ISO>', finishedAt: '<ISO>'
 };
 writeFileSync(RESULTS_FILE, JSON.stringify({ results, summary }, null, 2));
 console.log(`\n结果已写入 ${RESULTS_FILE}`);
@@ -389,6 +395,7 @@ console.log(`\n结果已写入 ${RESULTS_FILE}`);
 > - 凭证从本地 cache 读 + 本地轻量接口验证，**完全不依赖浏览器当前页面 origin**
 > - 同一次执行内 `credentialRefreshCount` 不得 > 1
 > - payload 从数据契约/接口定义/真实样例生成，并在脚本注释中标注字段来源
+> - **§5.22 分层耗时**：`batchDurationMs` 是 runner wall-clock（整个批次耗时），`requestDurationMs` 是单 HTTP 请求耗时。聚合合同套件场景（如 API-C01~C04 共享同一 pytest 批次）只记录 `batchReference` + `coveredTests`，**禁止用 `batchDurationMs / N` 均摊生成伪 `requestDurationMs`**
 
 ### 执行器降级方案（PowerShell batch runner .ps1）
 
@@ -445,18 +452,20 @@ $results | ConvertTo-Json -Depth 4
 | 接口测试执行器（普通） | 30s | **60s** | 抛错继续报告，不再重试 |
 | 接口测试执行器（复杂场景） | 60s | **120s** | 同上 |
 | 凭证获取（认证服务） | 5–10s | **15s** | 计 1 次重取；超过 1 次 → 🟡 WARN |
-| 单个 HTTP GET | < 1s | **8s** | durationMs > 10s → 🟡 SLOW |
-| 单个 HTTP POST/PUT | < 2s | **10s** | durationMs > 10s → 🟡 SLOW |
+| 单个 HTTP GET | < 1s | **8s** | requestDurationMs > 10s → 🟡 SLOW |
+| 单个 HTTP POST/PUT | < 2s | **10s** | requestDurationMs > 10s → 🟡 SLOW |
 | health probe | < 1s | **3s** | 计入启动状态机 |
 | 构建工具编译（如 Maven compile） | 10–60s | **180s** | 输出最后 30 行日志后停 |
 | 测试命令（如 Maven test） | 60–180s | **300s** | 输出失败用例后停 |
 | service start | 30–60s | **120s** | 4.1 启动状态机收敛 |
 
-**反馈规则**：
+**§5.22 分层耗时与反馈规则**：
 - 等待 > 10s 必须输出一次状态行（启动等待、执行器、构建工具均适用）
 - 不得静默等待
-- `durationMs > 10000` → 🟡 SLOW
-- `durationMs > 30000` → ❌ TIMEOUT_RISK
+- **`requestDurationMs`**（单 HTTP 请求耗时）：`> 10000` → 🟡 SLOW；`> 30000` → ❌ TIMEOUT_RISK
+- **`batchDurationMs`**（runner wall-clock，整个批次耗时）：`> 180000`（3 分钟）→ 🟡 BATCH_SLOW；`> 300000`（5 分钟）→ ❌ BATCH_TIMEOUT_RISK
+- **聚合合同套件场景**（如 API-C01~C04 共享同一 pytest 批次）：只记录 `batchReference` + `coveredTests`，**禁止用 `batchDurationMs / N` 均摊生成伪 `requestDurationMs`**
+- 报告必须明确标注是哪一层越界（batch-level 还是 request-level），不得混用
 - 请求慢不得忽略，必须说明是服务、网络、认证还是工具调用开销
 
 ## 认证凭证缓存与复用
@@ -638,17 +647,23 @@ TEST_<change-name>_<timestamp>_<short-random>
 | cleanup | cleaned=X / leftover=Y | 接口 /<resource> 不支持删除，2 条遗留 |
 
 ### 接口测试
-| # | 场景 | 方法 | URL | 预期 | 实际 | durationMs | 状态 |
+| # | 场景 | 方法 | URL | 预期 | 实际 | requestDurationMs | 状态 |
 |:--:|------|:----:|-----|------|------|----------:|:--:|
 | API-001 | 创建资源 | POST | /api/xxx | 200, code=0 | 200, code=0 | 245 | ✅ PASS |
 | API-002 | 参数校验 | POST | /api/xxx | code=xxx | code=xxx | 180 | ✅ PASS |
 | API-003 | 查询子规则 | GET | /api/xxx/{id} | 200 | — | — | 🟡 BLOCKED（setup.createRule 失败） |
 
-### 请求耗时统计
-| 场景 | 方法 | URL | durationMs | 状态 |
-|------|:----:|-----|----------:|:----:|
-| API-001 | POST | /api/xxx | 245 | ✅ |
-| API-010 | GET | /api/yyy | 12450 | 🟡SLOW |
+### 请求耗时统计（§5.22 分层耗时）
+| 层级 | 场景/批次 | 方法 | URL | 耗时 | 状态 |
+|------|----------|:----:|-----|----:|:----:|
+| request | API-001 | POST | /api/xxx | 245ms | ✅ |
+| request | API-010 | GET | /api/yyy | 12450ms | 🟡SLOW |
+| batch | pytest-batch-1 | — | — | 158000ms | 🟡BATCH_SLOW |
+
+> **§5.22 分层耗时说明**：
+> - `request` 层级记录单 HTTP 请求耗时（`requestDurationMs`）
+> - `batch` 层级记录 runner wall-clock（`batchDurationMs`），整个 pytest 批次耗时
+> - 聚合合同套件场景（如 API-C01~C04 共享同一批次）只记录一行 `batch` 条目 + `coveredTests` 列表，**不得均摊为 4 个伪 `request` 条目**
 
 ### 数据兼容
 | # | 场景 | 预期 | 实际 | 结果 |

package/harness/bundles/general/claude-code/protocols/submit-protocol.md

@@ -1,5 +1,5 @@
 ---
-description: harness-submit 的确定性提交协议。用于避免 AskUserQuestion 参数错误、中文 commit message 失败、Co-Authored-By 违规和先 commit 后 amend。
+description: harness-submit 的确定性提交协议。用于避免 blocking user confirmation 参数错误、中文 commit message 失败、Co-Authored-By 违规和先 commit 后 amend。
 ---
 
 # Submit Protocol

package/harness/bundles/general/claude-code/scripts/harness_deploy.py

@@ -26,6 +26,10 @@ if hasattr(sys.stderr, "reconfigure"):
     sys.stderr.reconfigure(encoding="utf-8")
 
 INCLUDE_RE = re.compile(r"<!--\s*@include\s+shared/([^\s]+)\s*-->")
+# retro §5.17: wiki links [[shared/xxx.md|alias]] reference shared fragments
+# that are inlined by expand_includes. After expansion, replace the wiki link
+# with the alias text (the shared content is already in the document).
+WIKI_LINK_RE = re.compile(r"\[\[shared/([^\]|]+)\|([^\]]*)\]\]")
 # §7.5: overlay anchors support both new section-id and legacy section:"title".
 OVERRIDE_RE = re.compile(r'<!--\s*@override\s+(?:section-id:"([^"]+)"|section:"([^"]+)")\s*-->')
 APPEND_RE = re.compile(r'<!--\s*@append-after\s+(?:section-id:"([^"]+)"|section:"([^"]+)")\s*-->')
@@ -148,6 +152,9 @@ def expand_includes(text: str, shared_dir: Path) -> str:
         return body
 
     expanded = INCLUDE_RE.sub(repl, text)
+    # §5.17: replace [[shared/xxx.md|alias]] wiki links with the alias text,
+    # since the shared content has been inlined by @include expansion above.
+    expanded = WIKI_LINK_RE.sub(lambda m: m.group(2), expanded)
     lines = [ln for ln in expanded.splitlines() if not FRAGMENT_HINT_RE.match(ln)]
     return "\n".join(lines) + ("\n" if text.endswith("\n") else "")
 

package/harness/bundles/general/claude-code/scripts/harness_gate.py

@@ -1201,6 +1201,70 @@ def _phase_event_exists(change_dir: Path, phase: str, type_: str, run_id: str) -
     )
 
 
+def _validate_scenario_coverage(change_dir: Path) -> dict[str, Any]:
+    """C9: validate all P0 scenarios have a ledger entry with scenarioIds covering them.
+
+    Reads meta/scenario-manifest.json and evidence/verification-ledger.json.
+    Returns ok=True when manifest missing (backward compat) or all P0 scenarios covered.
+    """
+    manifest_path = change_dir / "meta" / "scenario-manifest.json"
+    if not manifest_path.is_file():
+        return {"ok": True, "code": "MANIFEST_MISSING", "skipped": True}
+    try:
+        manifest = json.loads(manifest_path.read_text(encoding="utf-8-sig"))
+    except (OSError, json.JSONDecodeError) as exc:
+        return {
+            "ok": False,
+            "code": "SCENARIO_MANIFEST_INVALID",
+            "message": f"scenario-manifest.json unreadable: {exc}",
+        }
+    scenarios = manifest.get("scenarios") or []
+    p0_ids = {
+        str(s.get("id"))
+        for s in scenarios
+        if isinstance(s, dict)
+        and str(s.get("priority", "")).upper() == "P0"
+        and s.get("requiredEvidenceKind") == "ledger"
+    }
+    if not p0_ids:
+        return {"ok": True, "code": "NO_P0_SCENARIOS"}
+
+    ledger_path = change_dir / "evidence" / "verification-ledger.json"
+    if not ledger_path.is_file():
+        return {
+            "ok": False,
+            "code": "SCENARIO_COVERAGE_FAILED",
+            "message": "ledger missing; cannot verify P0 scenario coverage",
+            "missing": sorted(p0_ids),
+        }
+    try:
+        ledger = json.loads(ledger_path.read_text(encoding="utf-8-sig"))
+    except (OSError, json.JSONDecodeError) as exc:
+        return {
+            "ok": False,
+            "code": "SCENARIO_COVERAGE_FAILED",
+            "message": f"ledger unreadable: {exc}",
+            "missing": sorted(p0_ids),
+        }
+    covered: set[str] = set()
+    validations = ledger.get("validations") or {}
+    if isinstance(validations, dict):
+        for entry in validations.values():
+            if isinstance(entry, dict):
+                ids = entry.get("scenarioIds")
+                if isinstance(ids, list):
+                    covered.update(str(i) for i in ids)
+    missing = sorted(p0_ids - covered)
+    if missing:
+        return {
+            "ok": False,
+            "code": "SCENARIO_COVERAGE_FAILED",
+            "message": f"P0 scenarios without ledger entry: {', '.join(missing)}",
+            "missing": missing,
+        }
+    return {"ok": True, "code": "SCENARIO_COVERAGE_OK", "covered": sorted(covered & p0_ids)}
+
+
 def cmd_begin(args: argparse.Namespace) -> int:
     as_json = bool(args.json)
     project = hc.resolve_main_project_root()
@@ -1470,6 +1534,17 @@ def cmd_close(args: argparse.Namespace) -> int:
             extra={k: v for k, v in ledger_result.items() if k not in {"ok", "message", "code"}},
         )
 
+    # C9: scenario coverage check — all P0 scenarios must have a ledger entry.
+    if args.phase in {"run", "test"}:
+        coverage = _validate_scenario_coverage(change_dir)
+        if not coverage.get("ok"):
+            return emit_error(
+                str(coverage.get("code", "SCENARIO_COVERAGE_FAILED")),
+                str(coverage.get("message", "scenario coverage validation failed")),
+                as_json=as_json,
+                extra={k: v for k, v in coverage.items() if k not in {"ok", "message", "code"}},
+            )
+
     close_status = args.status
     close_code = "PHASE_CLOSED"
     if ledger_result.get("code") == "LEDGER_OK_DEGRADED":

package/harness/bundles/general/claude-code/scripts/harness_integration.py

@@ -1049,10 +1049,11 @@ class IntegrationTransaction:
                     heavy_removed.append(name)
                 except OSError as exc:
                     heavy_failures.append({"path": name, "error": str(exc)})
-        self.runner.run(
+        remove_result = self.runner.run(
             self.project_root, "worktree", "remove", "--force", str(resolved),
             check=False,
         )
+        remove_returncode = getattr(remove_result, "returncode", 0)
         listed = self.runner.text(self.project_root, "worktree", "list", "--porcelain")
         registered_after = {
             str(Path(line.removeprefix("worktree ")).resolve())
@@ -1061,7 +1062,8 @@ class IntegrationTransaction:
         }
         still_registered = str(resolved) in registered_after
         residual_paths: list[str] = []
-        if resolved.exists():
+        disk_path_present = resolved.exists()
+        if disk_path_present:
             if resolved.is_file():
                 residual_paths.append(resolved.name)
             else:
@@ -1075,6 +1077,27 @@ class IntegrationTransaction:
                             break
                     if len(residual_paths) >= 200:
                         break
+        # §5.30: Windows half-success — git worktree remove returned non-zero
+        # but registration is already deleted and disk path remains. Return a
+        # structured status distinct from generic CLEANUP_RESIDUAL so callers
+        # can apply the allowlisted residual cleaner.
+        if (
+            remove_returncode != 0
+            and not still_registered
+            and disk_path_present
+            and residual_paths
+            and not heavy_failures
+        ):
+            return {
+                "ok": False,
+                "code": "REGISTRATION_REMOVED_RESIDUAL_PRESENT",
+                "target": str(resolved),
+                "heavyRootsRemoved": heavy_removed,
+                "heavyRootFailures": heavy_failures,
+                "worktreeRegistered": False,
+                "diskPathPresent": True,
+                "residualPaths": residual_paths,
+            }
         ok = not heavy_failures and not still_registered and not residual_paths
         return {
             "ok": ok,
@@ -1083,6 +1106,7 @@ class IntegrationTransaction:
             "heavyRootsRemoved": heavy_removed,
             "heavyRootFailures": heavy_failures,
             "worktreeRegistered": still_registered,
+            "diskPathPresent": disk_path_present,
             "residualPaths": residual_paths,
         }
 
@@ -1266,6 +1290,31 @@ def cmd_status(args: argparse.Namespace) -> int:
     return _emit(_txn_from_args(args).status())
 
 
+def cmd_journal(args: argparse.Namespace) -> int:
+    """C5: journal 子命令 — 默认 compact (transactionId/currentStep/status)，--verbose 全量。"""
+    txn = _txn_from_args(args)
+    full = txn.status()
+    if getattr(args, "verbose", False):
+        return _emit(full)
+    # Derive currentStep: first non-DONE step, or last step name if all DONE.
+    current_step = None
+    for step in full.get("steps", []):
+        if step.get("status") != "DONE":
+            current_step = step.get("name")
+            break
+    if current_step is None and full.get("steps"):
+        current_step = full["steps"][-1].get("name")
+    overall_status = "DONE" if all(
+        step.get("status") == "DONE" for step in full.get("steps", [])
+    ) else "IN_PROGRESS"
+    compact = {
+        "transactionId": full.get("transactionId"),
+        "currentStep": current_step,
+        "status": overall_status,
+    }
+    return _emit(compact)
+
+
 def build_parser() -> argparse.ArgumentParser:
     parser = argparse.ArgumentParser(prog="harness_integration.py")
     sub = parser.add_subparsers(dest="command_name", required=True)
@@ -1287,6 +1336,19 @@ def build_parser() -> argparse.ArgumentParser:
     add("recover", cmd_recover)
     add("status", cmd_status)
 
+    p_journal = sub.add_parser("journal")
+    p_journal.add_argument("--change", required=True)
+    p_journal.add_argument("--run-id", required=True)
+    p_journal.add_argument("--target-branch", default="main")
+    p_journal.add_argument("--feature-branch", required=True)
+    p_journal.add_argument("--temp-root", required=True)
+    p_journal.add_argument(
+        "--verbose",
+        action="store_true",
+        help="emit full journal payload (default: compact transactionId/currentStep/status)",
+    )
+    p_journal.set_defaults(func=cmd_journal)
+
     p_verify = sub.add_parser("verify")
     p_verify.add_argument("--change", required=True)
     p_verify.add_argument("--run-id", required=True)

package/harness/bundles/general/claude-code/scripts/harness_ledger.py

@@ -100,6 +100,37 @@ def emit_json(payload: dict[str, Any], *, as_json: bool) -> None:
             sys.stdout.write(("ok" if ok else "error") + "\n")
 
 
+def _compact_record_payload(payload: dict[str, Any]) -> dict[str, Any]:
+    """C5: record compact — ok/action/verification/status only."""
+    return {
+        "ok": payload.get("ok", True),
+        "action": payload.get("action"),
+        "verification": payload.get("verification"),
+        "status": payload.get("status"),
+    }
+
+
+def _compact_can_reuse_payload(payload: dict[str, Any]) -> dict[str, Any]:
+    """C5: can-reuse compact — ok/reuse/code only."""
+    return {
+        "ok": payload.get("ok", True),
+        "reuse": payload.get("reuse"),
+        "code": payload.get("code"),
+    }
+
+
+def emit_compact_or_verbose(
+    payload: dict[str, Any],
+    *,
+    as_json: bool,
+    verbose: bool,
+    compact_fn,
+) -> None:
+    """Emit compact payload by default; full payload when --verbose."""
+    out = payload if verbose else compact_fn(payload)
+    emit_json(out, as_json=as_json)
+
+
 def emit_error(message: str, *, as_json: bool, code: int = 1) -> int:
     payload = {"ok": False, "error": message}
     if as_json:
@@ -1227,6 +1258,7 @@ def cmd_hash(args: argparse.Namespace) -> int:
 
 def cmd_can_reuse(args: argparse.Namespace) -> int:
     as_json = bool(args.json)
+    verbose = bool(getattr(args, "verbose", False))
     verification = args.verification
     if verification not in VERIFICATIONS:
         return emit_error(
@@ -1253,7 +1285,10 @@ def cmd_can_reuse(args: argparse.Namespace) -> int:
                 "verification": verification,
                 "detail": err,
             }
-            emit_json(payload, as_json=as_json)
+            emit_compact_or_verbose(
+                payload, as_json=as_json, verbose=verbose,
+                compact_fn=_compact_can_reuse_payload,
+            )
             return 0
         files = resolve_input_files(resolved_files, project_root)
     if not files:
@@ -1275,12 +1310,16 @@ def cmd_can_reuse(args: argparse.Namespace) -> int:
     except (OSError, ValueError, json.JSONDecodeError) as exc:
         return emit_error(f"can-reuse failed: {exc}", as_json=as_json)
 
-    emit_json(payload, as_json=as_json)
+    emit_compact_or_verbose(
+        payload, as_json=as_json, verbose=verbose,
+        compact_fn=_compact_can_reuse_payload,
+    )
     return 0
 
 
 def cmd_record(args: argparse.Namespace) -> int:
     as_json = bool(args.json)
+    verbose = bool(getattr(args, "verbose", False))
     change_dir = resolve_path(args.change_dir)
     verification = args.verification
     if verification not in VERIFICATIONS:
@@ -1421,6 +1460,16 @@ def cmd_record(args: argparse.Namespace) -> int:
             if _nonempty_str(getattr(args, "tests_reused_from", None)):
                 entry["testsReusedFrom"] = str(args.tests_reused_from).strip()
 
+        # C9: bind scenario IDs from --scenario-ids to this ledger entry.
+        scenario_ids_raw = getattr(args, "scenario_ids", None)
+        if _nonempty_str(scenario_ids_raw):
+            ids = [s.strip() for s in str(scenario_ids_raw).split(",") if s.strip()]
+            if ids:
+                entry["scenarioIds"] = ids
+        elif "scenarioIds" in entry:
+            # Fresh record without scenario-ids must not keep stale ids from prev.
+            entry.pop("scenarioIds", None)
+
         ledger["validations"][verification] = entry
         if "changeName" not in ledger:
             ledger["changeName"] = change_dir.name
@@ -1487,7 +1536,10 @@ def cmd_record(args: argparse.Namespace) -> int:
         "diffHash": ledger.get("diffHash"),
         "resolvedProjectRoot": str(project_root) if project_root else None,
     }
-    emit_json(payload, as_json=as_json)
+    emit_compact_or_verbose(
+        payload, as_json=as_json, verbose=verbose,
+        compact_fn=_compact_record_payload,
+    )
     return 0
 
 
@@ -1589,6 +1641,11 @@ def build_parser() -> argparse.ArgumentParser:
         default=None,
         help="optional environment hash to compare against ledger entry (UT-017)",
     )
+    p_reuse.add_argument(
+        "--verbose",
+        action="store_true",
+        help="emit full payload (default: compact ok/reuse/code)",
+    )
     p_reuse.set_defaults(func=cmd_can_reuse)
 
     p_record = sub.add_parser("record", parents=[shared_json], help="write validation result into ledger")
@@ -1669,6 +1726,16 @@ def build_parser() -> argparse.ArgumentParser:
         default=None,
         help="ledger v3: scope reason (required when --applicability NOT_APPLICABLE)",
     )
+    p_record.add_argument(
+        "--scenario-ids",
+        default=None,
+        help="comma-separated scenario IDs from scenario-manifest.json to bind to this entry",
+    )
+    p_record.add_argument(
+        "--verbose",
+        action="store_true",
+        help="emit full payload (default: compact ok/action/verification/status)",
+    )
     p_record.set_defaults(func=cmd_record)
 
     p_diff = sub.add_parser(

package/harness/bundles/general/claude-code/scripts/harness_paths.py

@@ -83,6 +83,23 @@ def _git_common_dir(cwd: Path) -> Path | None:
     return common
 
 
+def common_root(project_root: Path) -> Path:
+    """Resolve the shared "common" root for a project.
+
+    For a linked worktree, ``git rev-parse --git-common-dir`` points at the
+    main repository's ``.git`` directory; its parent is the main project root.
+    For the main worktree, the common dir is ``<root>/.git`` so the parent is
+    the project root itself. Falls back to the resolved input when not in a
+    git repo (e.g. exploratory scratch dirs).
+    """
+    cwd = Path(project_root).resolve()
+    common = _git_common_dir(cwd)
+    if common is None:
+        return cwd
+    # common dir is <root>/.git — its parent is the main project root.
+    return common.parent.resolve()
+
+
 def _root_commit(cwd: Path) -> str | None:
     raw = _git_text(cwd, "rev-list", "--max-parents=0", "HEAD")
     if not raw: