superlab 0.1.14 → 0.1.15

package/README.md

@@ -179,7 +179,7 @@ superlab auto stop
 
 - `run` and `iterate` must change persistent outputs under `results_root`
 - `review` must update canonical review context
-- `report` must write `<deliverables_root>/report.md`
+- `report` must write `<deliverables_root>/report.md` and `<deliverables_root>/main-tables.md`
 - `write` must produce LaTeX output under `<deliverables_root>/paper/`
 - a successful promotion must write back into `.lab/context/data-decisions.md`, `.lab/context/decisions.md`, `.lab/context/state.md`, and `.lab/context/session-brief.md`
 - every run must end with `.lab/context/auto-outcome.md`, including why it stopped, whether the terminal goal was reached, and which artifact is the final outcome
@@ -246,7 +246,7 @@ Stages should follow that file rather than guess language locally.
 - Python 3.10+
 - Git
 
-`/lab:write` ships with vendored paper-writing references under the installed `lab` skill, so it does not depend on an extra runtime skill installation.
+`/lab:write` ships with vendored paper-writing references and the upstream example bank for `abstract`, `introduction`, and `method` under the installed `lab` skill, so it does not depend on an extra runtime skill installation.
 
 ## Command Set
 
@@ -294,6 +294,7 @@ See the source command docs in [commands/codex/lab.md](/Users/zhouhao119/coding/
 `/lab:write` writes final manuscript output under the configured `deliverables_root` (default: `docs/research`):
 
 - `docs/research/report.md`
+- `docs/research/main-tables.md`
 - `docs/research/paper/main.tex`
 - `docs/research/paper/sections/*.tex`
 
@@ -303,7 +304,10 @@ Internal writing-control artifacts stay under:
 - `.lab/writing/plan.md`
 - `.lab/writing/iterations/*.md`
 
-If `paper_template_root` is configured, `/lab:write` should inspect that template directory first and align drafting to it. If no template is configured, it falls back to the managed default LaTeX scaffold.
+If `paper_template_root` is configured, `/lab:write` should inspect that template directory first and align drafting to it.
+If no template is configured, the first manuscript-writing round should ask once whether to continue with the managed default LaTeX scaffold or attach a template directory first.
+If the user approves the default scaffold, persist that choice in `.lab/config/workflow.json` and stop asking on ordinary rounds.
+At the final export or final-draft boundary, if the project is still on the default scaffold and no attached template exists, ask one final reminder question before finalizing.
 
 `/lab` treats `.lab/` as the workflow-control layer only. Durable outputs should use natural project roots:
 

package/README.zh-CN.md

@@ -177,7 +177,7 @@ superlab auto stop
 
 - `run` 和 `iterate` 必须更新 `results_root` 下的持久输出
 - `review` 必须更新规范的审查上下文
-- `report` 必须写出 `<deliverables_root>/report.md`
+- `report` 必须写出 `<deliverables_root>/report.md` 和 `<deliverables_root>/main-tables.md`
 - `write` 必须写出 `<deliverables_root>/paper/` 下的 LaTeX 论文产物
 - promotion 成功后必须写回 `.lab/context/data-decisions.md`、`.lab/context/decisions.md`、`.lab/context/state.md` 和 `.lab/context/session-brief.md`
 - 每次运行都必须写出 `.lab/context/auto-outcome.md`，记录为什么停止、是否达到终止目标，以及哪一个工件是最终结果
@@ -244,7 +244,7 @@ superlab init --lang en
 - Python 3.10+
 - Git
 
-`/lab:write` 自带 vendored 的 paper-writing 章节参考，不再依赖额外安装一个运行时写作 skill。
+`/lab:write` 自带 vendored 的 paper-writing 章节参考，以及 `abstract`、`introduction`、`method` 对应的 upstream example bank，不再依赖额外安装一个运行时写作 skill。
 
 ## 命令集合
 
@@ -279,6 +279,7 @@ superlab init --lang en
 `/lab:write` 会把最终可交付物写到 `deliverables_root` 指定的目录，默认是 `docs/research`：
 
 - `docs/research/report.md`
+- `docs/research/main-tables.md`
 - `docs/research/paper/main.tex`
 - `docs/research/paper/sections/*.tex`
 
@@ -288,7 +289,10 @@ superlab init --lang en
 - `.lab/writing/plan.md`
 - `.lab/writing/iterations/*.md`
 
-如果配置了 `paper_template_root`，`/lab:write` 应先检查该模板目录并按其结构写作；如果没有配置模板，就回退到内置的默认 LaTeX scaffold。
+如果配置了 `paper_template_root`，`/lab:write` 应先检查该模板目录并按其结构写作。
+如果没有配置模板，第一次进入论文 `.tex` 写作时应先追问一次：继续使用内置默认 LaTeX scaffold，还是先接入模板目录。
+如果用户确认先用默认 scaffold，就把这个决定持久化到 `.lab/config/workflow.json`，后续普通轮次不再重复追问。
+但在最终导出或最终定稿节点，如果项目仍在使用默认 scaffold 且没有接入模板，应再提醒一次，给用户最后切换模板的机会。
 
 `/lab` 把 `.lab/` 视为工作流控制层，不是正式结果目录。持久输出应按自然根目录放置：
 

package/bin/superlab.cjs

@@ -603,6 +603,8 @@ function attachPaperTemplate({ targetDir, templatePath }) {
   }
 
   config.paper_template_root = storedProjectPath(targetDir, normalizedTemplatePath);
+  config.paper_template_decision = "attached-template";
+  config.paper_template_final_reminder_acknowledged = true;
   writeWorkflowConfig(targetDir, config);
   return {
     storedPath: config.paper_template_root,
@@ -660,6 +662,33 @@ function validateWorkflowConfig(config) {
       issues.push("paper_template_root must not point inside .lab");
     }
   }
+  const validTemplateDecisions = new Set(["unconfirmed", "default-scaffold", "attached-template"]);
+  if (
+    config.paper_template_decision !== undefined &&
+    !validTemplateDecisions.has(config.paper_template_decision)
+  ) {
+    issues.push("invalid paper_template_decision");
+  }
+  if (
+    config.paper_template_final_reminder_acknowledged !== undefined &&
+    typeof config.paper_template_final_reminder_acknowledged !== "boolean"
+  ) {
+    issues.push("invalid paper_template_final_reminder_acknowledged");
+  }
+  if (
+    typeof config.paper_template_root === "string" &&
+    config.paper_template_root.trim() !== "" &&
+    config.paper_template_decision === "default-scaffold"
+  ) {
+    issues.push("paper_template_decision conflicts with configured paper_template_root");
+  }
+  if (
+    typeof config.paper_template_root === "string" &&
+    config.paper_template_root.trim() === "" &&
+    config.paper_template_decision === "attached-template"
+  ) {
+    issues.push("paper_template_decision requires a configured paper_template_root");
+  }
   return issues;
 }
 
@@ -922,6 +951,15 @@ async function main() {
       console.log(`stages executed: ${result.executedStages.join(", ")}`);
       console.log(`goal type: ${result.outcome.goalType}`);
       console.log(`goal target: ${result.outcome.goalTarget}`);
+      if (result.outcome.primaryMetrics) {
+        console.log(`primary metrics: ${result.outcome.primaryMetrics}`);
+      }
+      if (result.outcome.secondaryMetrics) {
+        console.log(`secondary metrics: ${result.outcome.secondaryMetrics}`);
+      }
+      if (result.outcome.requiredTerminalEvidence) {
+        console.log(`required terminal evidence: ${result.outcome.requiredTerminalEvidence}`);
+      }
       if (result.outcome.experimentLadder) {
         console.log(`experiment ladder: ${result.outcome.experimentLadder}`);
       }

package/lib/auto_contracts.cjs

@@ -285,7 +285,10 @@ function stageContractSnapshot(targetDir, stage) {
     run: [resultsRoot],
     iterate: [resultsRoot],
     review: REVIEW_CONTEXT_FILES.map((relativePath) => path.resolve(targetDir, relativePath)),
-    report: [path.join(deliverablesRoot, "report.md")],
+    report: [
+      path.join(deliverablesRoot, "report.md"),
+      path.join(deliverablesRoot, "main-tables.md"),
+    ],
     write: [
       path.join(deliverablesRoot, "paper", "main.tex"),
       path.join(deliverablesRoot, "paper", "sections"),
@@ -322,8 +325,9 @@ function verifyStageContract({ stage, snapshot }) {
   }
 
   if (stage === "report") {
-    if (changedPaths.length === 0) {
-      throw new Error("report stage did not produce the deliverable report.md under deliverables_root");
+    const missing = Array.from(snapshot.keys()).filter((absolutePath) => !changedPaths.includes(absolutePath));
+    if (missing.length > 0) {
+      throw new Error("report stage did not produce the deliverable report.md and main-tables.md under deliverables_root");
     }
     return;
   }

package/lib/auto_runner.cjs

@@ -264,6 +264,18 @@ async function startAutoMode({ targetDir, now = new Date() }) {
   let promotionApplied = false;
   let stopReason = "";
   let finalRung = "";
+  const outcomeProtocolFields = {
+    primaryMetrics: evalProtocol.primaryMetrics,
+    secondaryMetrics: evalProtocol.secondaryMetrics,
+    requiredTerminalEvidence: evalProtocol.requiredTerminalEvidence,
+    experimentLadder: evalProtocol.experimentLadder,
+    metricGlossary: evalProtocol.metricGlossary,
+    metricSourcePapers: evalProtocol.metricSourcePapers,
+    metricImplementationSource: evalProtocol.metricImplementationSource,
+    comparisonSourcePapers: evalProtocol.comparisonSourcePapers,
+    comparisonImplementationSource: evalProtocol.comparisonImplementationSource,
+    deviationFromOriginalImplementation: evalProtocol.deviationFromOriginalImplementation,
+  };
 
   const writeRunningStatus = (overrides = {}) => {
     currentStatus = {
@@ -287,13 +299,7 @@ async function startAutoMode({ targetDir, now = new Date() }) {
       targetDir,
       {
         objective: mode.objective,
-        experimentLadder: evalProtocol.experimentLadder,
-        metricGlossary: evalProtocol.metricGlossary,
-        metricSourcePapers: evalProtocol.metricSourcePapers,
-        metricImplementationSource: evalProtocol.metricImplementationSource,
-        comparisonSourcePapers: evalProtocol.comparisonSourcePapers,
-        comparisonImplementationSource: evalProtocol.comparisonImplementationSource,
-        deviationFromOriginalImplementation: evalProtocol.deviationFromOriginalImplementation,
+        ...outcomeProtocolFields,
         terminalGoalType: mode.terminalGoalType,
         terminalGoalTarget: mode.terminalGoalTarget,
         requiredTerminalArtifact: mode.requiredTerminalArtifact,
@@ -597,8 +603,7 @@ async function startAutoMode({ targetDir, now = new Date() }) {
       targetDir,
       {
         objective: mode.objective,
-        experimentLadder: evalProtocol.experimentLadder,
-        metricGlossary: evalProtocol.metricGlossary,
+        ...outcomeProtocolFields,
         terminalGoalType: mode.terminalGoalType,
         terminalGoalTarget: mode.terminalGoalTarget,
         requiredTerminalArtifact: mode.requiredTerminalArtifact,
@@ -620,14 +625,8 @@ async function startAutoMode({ targetDir, now = new Date() }) {
       mode,
       status: currentStatus,
       executedStages,
-      outcome: {
-        experimentLadder: evalProtocol.experimentLadder,
-        metricGlossary: evalProtocol.metricGlossary,
-        metricSourcePapers: evalProtocol.metricSourcePapers,
-        metricImplementationSource: evalProtocol.metricImplementationSource,
-        comparisonSourcePapers: evalProtocol.comparisonSourcePapers,
-        comparisonImplementationSource: evalProtocol.comparisonImplementationSource,
-        deviationFromOriginalImplementation: evalProtocol.deviationFromOriginalImplementation,
+        outcome: {
+        ...outcomeProtocolFields,
         goalType: mode.terminalGoalType,
         goalTarget: mode.terminalGoalTarget,
         goalReached: false,
@@ -656,13 +655,7 @@ async function startAutoMode({ targetDir, now = new Date() }) {
           targetDir,
           {
             objective: mode.objective,
-            experimentLadder: evalProtocol.experimentLadder,
-            metricGlossary: evalProtocol.metricGlossary,
-            metricSourcePapers: evalProtocol.metricSourcePapers,
-            metricImplementationSource: evalProtocol.metricImplementationSource,
-            comparisonSourcePapers: evalProtocol.comparisonSourcePapers,
-            comparisonImplementationSource: evalProtocol.comparisonImplementationSource,
-            deviationFromOriginalImplementation: evalProtocol.deviationFromOriginalImplementation,
+            ...outcomeProtocolFields,
             terminalGoalType: mode.terminalGoalType,
             terminalGoalTarget: mode.terminalGoalTarget,
             requiredTerminalArtifact: mode.requiredTerminalArtifact,
@@ -685,13 +678,7 @@ async function startAutoMode({ targetDir, now = new Date() }) {
           status: currentStatus,
           executedStages,
           outcome: {
-            experimentLadder: evalProtocol.experimentLadder,
-            metricGlossary: evalProtocol.metricGlossary,
-            metricSourcePapers: evalProtocol.metricSourcePapers,
-            metricImplementationSource: evalProtocol.metricImplementationSource,
-            comparisonSourcePapers: evalProtocol.comparisonSourcePapers,
-            comparisonImplementationSource: evalProtocol.comparisonImplementationSource,
-            deviationFromOriginalImplementation: evalProtocol.deviationFromOriginalImplementation,
+            ...outcomeProtocolFields,
             goalType: mode.terminalGoalType,
             goalTarget: mode.terminalGoalTarget,
             goalReached: false,
@@ -722,13 +709,7 @@ async function startAutoMode({ targetDir, now = new Date() }) {
     targetDir,
     {
       objective: mode.objective,
-      experimentLadder: evalProtocol.experimentLadder,
-      metricGlossary: evalProtocol.metricGlossary,
-      metricSourcePapers: evalProtocol.metricSourcePapers,
-      metricImplementationSource: evalProtocol.metricImplementationSource,
-      comparisonSourcePapers: evalProtocol.comparisonSourcePapers,
-      comparisonImplementationSource: evalProtocol.comparisonImplementationSource,
-      deviationFromOriginalImplementation: evalProtocol.deviationFromOriginalImplementation,
+      ...outcomeProtocolFields,
       terminalGoalType: mode.terminalGoalType,
       terminalGoalTarget: mode.terminalGoalTarget,
       requiredTerminalArtifact: mode.requiredTerminalArtifact,
@@ -751,13 +732,7 @@ async function startAutoMode({ targetDir, now = new Date() }) {
     status: currentStatus,
     executedStages,
     outcome: {
-      experimentLadder: evalProtocol.experimentLadder,
-      metricGlossary: evalProtocol.metricGlossary,
-      metricSourcePapers: evalProtocol.metricSourcePapers,
-      metricImplementationSource: evalProtocol.metricImplementationSource,
-      comparisonSourcePapers: evalProtocol.comparisonSourcePapers,
-      comparisonImplementationSource: evalProtocol.comparisonImplementationSource,
-      deviationFromOriginalImplementation: evalProtocol.deviationFromOriginalImplementation,
+      ...outcomeProtocolFields,
       goalType: mode.terminalGoalType,
       goalTarget: mode.terminalGoalTarget,
       goalReached: successReached,
@@ -774,6 +749,18 @@ function stopAutoMode({ targetDir, now = new Date() }) {
   const mode = parseAutoMode(targetDir);
   const evalProtocol = parseEvalProtocol(targetDir);
   const lang = readWorkflowLanguage(targetDir);
+  const outcomeProtocolFields = {
+    primaryMetrics: evalProtocol.primaryMetrics,
+    secondaryMetrics: evalProtocol.secondaryMetrics,
+    requiredTerminalEvidence: evalProtocol.requiredTerminalEvidence,
+    experimentLadder: evalProtocol.experimentLadder,
+    metricGlossary: evalProtocol.metricGlossary,
+    metricSourcePapers: evalProtocol.metricSourcePapers,
+    metricImplementationSource: evalProtocol.metricImplementationSource,
+    comparisonSourcePapers: evalProtocol.comparisonSourcePapers,
+    comparisonImplementationSource: evalProtocol.comparisonImplementationSource,
+    deviationFromOriginalImplementation: evalProtocol.deviationFromOriginalImplementation,
+  };
   const status = {
     ...existing,
     status: "stopped",
@@ -785,13 +772,7 @@ function stopAutoMode({ targetDir, now = new Date() }) {
     targetDir,
     {
       objective: mode.objective,
-      experimentLadder: evalProtocol.experimentLadder,
-      metricGlossary: evalProtocol.metricGlossary,
-      metricSourcePapers: evalProtocol.metricSourcePapers,
-      metricImplementationSource: evalProtocol.metricImplementationSource,
-      comparisonSourcePapers: evalProtocol.comparisonSourcePapers,
-      comparisonImplementationSource: evalProtocol.comparisonImplementationSource,
-      deviationFromOriginalImplementation: evalProtocol.deviationFromOriginalImplementation,
+      ...outcomeProtocolFields,
       terminalGoalType: mode.terminalGoalType,
       terminalGoalTarget: mode.terminalGoalTarget,
       requiredTerminalArtifact: mode.requiredTerminalArtifact,

package/lib/auto_state.cjs

@@ -140,30 +140,33 @@ function renderAutoOutcome(outcome, { lang = "en" } = {}) {
 
 ## 目标
 
-- Objective: ${outcome.objective || ""}
-- Experiment ladder: ${outcome.experimentLadder || ""}
-- Metric glossary: ${outcome.metricGlossary || ""}
-- Metric source papers: ${outcome.metricSourcePapers || ""}
-- Metric implementation source: ${outcome.metricImplementationSource || ""}
-- Comparison source papers: ${outcome.comparisonSourcePapers || ""}
-- Comparison implementation source: ${outcome.comparisonImplementationSource || ""}
-- Deviation from original implementation: ${outcome.deviationFromOriginalImplementation || ""}
-- Terminal goal type: ${outcome.terminalGoalType || ""}
-- Terminal goal target: ${outcome.terminalGoalTarget || ""}
-- Required terminal artifact: ${outcome.requiredTerminalArtifact || ""}
+- 目标: ${outcome.objective || ""}
+- 主指标: ${outcome.primaryMetrics || ""}
+- 次级指标: ${outcome.secondaryMetrics || ""}
+- 必要终局证据: ${outcome.requiredTerminalEvidence || ""}
+- 实验阶梯: ${outcome.experimentLadder || ""}
+- 指标释义: ${outcome.metricGlossary || ""}
+- 指标来源论文: ${outcome.metricSourcePapers || ""}
+- 指标实现来源: ${outcome.metricImplementationSource || ""}
+- 对比方法来源论文: ${outcome.comparisonSourcePapers || ""}
+- 对比方法实现来源: ${outcome.comparisonImplementationSource || ""}
+- 与原始实现的偏差: ${outcome.deviationFromOriginalImplementation || ""}
+- 终止目标类型: ${outcome.terminalGoalType || ""}
+- 终止目标目标值: ${outcome.terminalGoalTarget || ""}
+- 必要终止工件: ${outcome.requiredTerminalArtifact || ""}
 
 ## 结果
 
-- Status: ${outcome.status || ""}
-- Goal reached: ${outcome.goalReached ? "yes" : "no"}
-- Stop reason: ${outcome.stopReason || ""}
-- Promotion applied: ${outcome.promotionApplied ? "yes" : "no"}
-- Final artifact: ${outcome.finalArtifact || ""}
-- Final rung: ${outcome.finalRung || ""}
-- Executed stages: ${outcome.executedStages || ""}
-- Iterations completed: ${outcome.iterationsCompleted || "0"}
-- Started at: ${outcome.startedAt || ""}
-- Finished at: ${outcome.finishedAt || ""}
+- 状态: ${outcome.status || ""}
+- 目标是否达成: ${outcome.goalReached ? "是" : "否"}
+- 停止原因: ${outcome.stopReason || ""}
+- 是否已升格: ${outcome.promotionApplied ? "是" : "否"}
+- 最终工件: ${outcome.finalArtifact || ""}
+- 最终 rung: ${outcome.finalRung || ""}
+- 已执行阶段: ${outcome.executedStages || ""}
+- 已完成迭代数: ${outcome.iterationsCompleted || "0"}
+- 开始时间: ${outcome.startedAt || ""}
+- 结束时间: ${outcome.finishedAt || ""}
 `;
   }
 
@@ -172,6 +175,9 @@ function renderAutoOutcome(outcome, { lang = "en" } = {}) {
 ## Goal
 
 - Objective: ${outcome.objective || ""}
+- Primary metrics: ${outcome.primaryMetrics || ""}
+- Secondary metrics: ${outcome.secondaryMetrics || ""}
+- Required terminal evidence: ${outcome.requiredTerminalEvidence || ""}
 - Experiment ladder: ${outcome.experimentLadder || ""}
 - Metric glossary: ${outcome.metricGlossary || ""}
 - Metric source papers: ${outcome.metricSourcePapers || ""}

package/lib/context.cjs

@@ -96,6 +96,9 @@ function renderSummary(lang, data) {
 - Auto final artifact: ${data.autoFinalArtifact || "待补充"}
 - Auto final rung: ${data.autoFinalRung || "待补充"}
 - Eval objective: ${data.evalObjective || "待补充"}
+- Primary metrics: ${data.evalPrimaryMetrics || "待补充"}
+- Secondary metrics: ${data.evalSecondaryMetrics || "待补充"}
+- Required terminal evidence: ${data.evalRequiredTerminalEvidence || "待补充"}
 - Table plan: ${data.evalTablePlan || "待补充"}
 - Metric glossary: ${data.evalMetricGlossary || "待补充"}
 - Metric source papers: ${data.evalMetricSourcePapers || "待补充"}
@@ -148,6 +151,9 @@ function renderSummary(lang, data) {
 - Auto final artifact: ${data.autoFinalArtifact || "TBD"}
 - Auto final rung: ${data.autoFinalRung || "TBD"}
 - Eval objective: ${data.evalObjective || "TBD"}
+- Primary metrics: ${data.evalPrimaryMetrics || "TBD"}
+- Secondary metrics: ${data.evalSecondaryMetrics || "TBD"}
+- Required terminal evidence: ${data.evalRequiredTerminalEvidence || "TBD"}
 - Table plan: ${data.evalTablePlan || "TBD"}
 - Metric glossary: ${data.evalMetricGlossary || "TBD"}
 - Metric source papers: ${data.evalMetricSourcePapers || "TBD"}
@@ -255,6 +261,9 @@ ${data.problem || "待补充"}
 - Auto final artifact: ${data.autoFinalArtifact || "待补充"}
 - Auto final rung: ${data.autoFinalRung || "待补充"}
 - Eval objective: ${data.evalObjective || "待补充"}
+- Primary metrics: ${data.evalPrimaryMetrics || "待补充"}
+- Secondary metrics: ${data.evalSecondaryMetrics || "待补充"}
+- Required terminal evidence: ${data.evalRequiredTerminalEvidence || "待补充"}
 - Table plan: ${data.evalTablePlan || "待补充"}
 - Metric glossary: ${data.evalMetricGlossary || "待补充"}
 - Metric source papers: ${data.evalMetricSourcePapers || "待补充"}
@@ -318,6 +327,9 @@ ${data.problem || "TBD"}
 - Auto final artifact: ${data.autoFinalArtifact || "TBD"}
 - Auto final rung: ${data.autoFinalRung || "TBD"}
 - Eval objective: ${data.evalObjective || "TBD"}
+- Primary metrics: ${data.evalPrimaryMetrics || "TBD"}
+- Secondary metrics: ${data.evalSecondaryMetrics || "TBD"}
+- Required terminal evidence: ${data.evalRequiredTerminalEvidence || "TBD"}
 - Table plan: ${data.evalTablePlan || "TBD"}
 - Metric glossary: ${data.evalMetricGlossary || "TBD"}
 - Metric source papers: ${data.evalMetricSourcePapers || "TBD"}
@@ -569,6 +581,9 @@ function buildContextSnapshot(targetDir) {
     autoFinalArtifact: extractValue(autoOutcome, ["Final artifact", "最终工件"]),
     autoFinalRung: extractValue(autoOutcome, ["Final rung", "最终 rung"]),
     evalObjective: evalProtocol.primaryEvaluationObjective,
+    evalPrimaryMetrics: evalProtocol.primaryMetrics,
+    evalSecondaryMetrics: evalProtocol.secondaryMetrics,
+    evalRequiredTerminalEvidence: evalProtocol.requiredTerminalEvidence,
     evalTablePlan: evalProtocol.tablePlan,
     evalMetricGlossary: evalProtocol.metricGlossary,
     evalMetricSourcePapers: evalProtocol.metricSourcePapers,

package/lib/i18n.cjs

@@ -61,7 +61,7 @@ const ZH_CONTENT = {
   [path.join(".codex", "prompts", "lab-write.md")]: codexPrompt(
     "把验证过的研究工件转成论文 section，并按小步方式修订",
     "section or writing target",
-    "使用已安装的 `lab` 技能：`.codex/skills/lab/SKILL.md`。\n\n立刻针对用户当前给出的参数执行 `/lab:write`，不要只推荐别的 `/lab` 阶段。只有在缺少阻塞性前提时，才明确指出缺什么，并且一次最多追问一个问题。\n\n本命令运行 `/lab:write` 阶段。它必须先有来自 `/lab:framing` 的已批准 framing artifact，再读取 `.codex/skills/lab/references/paper-writing/` 下与当前 section 对应的参考文件，并结合 `paper-review.md` 与 `does-my-writing-flow-source.md`，先写 mini-outline，再只修改一个 section。"
+    "使用已安装的 `lab` 技能：`.codex/skills/lab/SKILL.md`。\n\n立刻针对用户当前给出的参数执行 `/lab:write`，不要只推荐别的 `/lab` 阶段。只有在缺少阻塞性前提时，才明确指出缺什么，并且一次最多追问一个问题。\n\n本命令运行 `/lab:write` 阶段。它必须先有来自 `/lab:framing` 的已批准 framing artifact，再读取 `.codex/skills/lab/references/paper-writing/` 下与当前 section 对应的参考文件；如果当前是 `abstract`、`introduction` 或 `method`，还必须继续读取 `.codex/skills/lab/references/paper-writing/examples/index.md`、对应的 examples index，以及 1-2 个具体 example 文件。然后结合 `paper-review.md` 与 `does-my-writing-flow-source.md`，先写 mini-outline，再只修改一个 section。第一次进入会产出论文 `.tex` 的路径时，如果 `paper_template_root` 为空，必须先问一次：继续使用默认 LaTeX scaffold，还是先接入模板目录。"
   ),
   [path.join(".claude", "commands", "lab", "idea.md")]: claudeCommand(
     "LAB: Idea",
@@ -109,7 +109,7 @@ const ZH_CONTENT = {
     "LAB: Write",
     "把验证过的研究工件转成论文 section，并按小步方式修订",
     "workflow, research, writing",
-    "使用已安装的 `lab` 技能：`.claude/skills/lab/SKILL.md`。\n\n立刻针对用户当前给出的参数执行 `/lab:write`，不要只推荐别的 `/lab` 阶段。只有在缺少阻塞性前提时，才明确指出缺什么，并且一次最多追问一个问题。\n\n本命令运行 `/lab:write` 阶段。它必须先有来自 `/lab:framing` 的已批准 framing artifact，再读取 `.claude/skills/lab/references/paper-writing/` 下与当前 section 对应的参考文件，并结合 `paper-review.md` 与 `does-my-writing-flow-source.md`，先写 mini-outline，再只修改一个 section。"
+    "使用已安装的 `lab` 技能：`.claude/skills/lab/SKILL.md`。\n\n立刻针对用户当前给出的参数执行 `/lab:write`，不要只推荐别的 `/lab` 阶段。只有在缺少阻塞性前提时，才明确指出缺什么，并且一次最多追问一个问题。\n\n本命令运行 `/lab:write` 阶段。它必须先有来自 `/lab:framing` 的已批准 framing artifact，再读取 `.claude/skills/lab/references/paper-writing/` 下与当前 section 对应的参考文件；如果当前是 `abstract`、`introduction` 或 `method`，还必须继续读取 `.claude/skills/lab/references/paper-writing/examples/index.md`、对应的 examples index，以及 1-2 个具体 example 文件。然后结合 `paper-review.md` 与 `does-my-writing-flow-source.md`，先写 mini-outline，再只修改一个 section。第一次进入会产出论文 `.tex` 的路径时，如果 `paper_template_root` 为空，必须先问一次：继续使用默认 LaTeX scaffold，还是先接入模板目录。"
   ),
 };
 
@@ -291,8 +291,10 @@ const ZH_SKILL_FILES = {
 ## 必要输出
 
 - 方法概述
+- 选定指标摘要
 - 实验设置
 - 已验证主结果
+- 位于 \`<deliverables_root>/main-tables.md\` 的受管主表工件
 - 消融
 - 失败尝试
 - 局限性
@@ -317,13 +319,15 @@ const ZH_SKILL_FILES = {
 - 每个主要 claim 都要指向已记录的 summary 或 iteration artifact。
 - 主表结构、gate 和最终结果 framing 必须对齐已批准的评估协议。
 - 不要凭记忆重述指标定义、baseline 行为或对比方法实现；直接引用评估协议里记录的来源。
+- 必须把已批准的主指标、次级指标和必要终局证据明确写进 \`report.md\` 与受管的 \`main-tables.md\`。
 - 如果报告依赖了对原始指标或原始实现的偏差，必须明确写出这个偏差。
+- 如果 workflow language 是中文，\`report.md\` 和 \`<deliverables_root>/main-tables.md\` 也应使用中文，除非文件路径、代码标识符或字面指标名必须保持原样。
 - 解释优先保守，不要写成营销文案。
 - 要给 \`/lab:write\` 留下清晰 handoff，尤其是 section draft 可以直接引用的证据链接。
 
 ## 交互约束
 
-- 开始前先简洁说明：campaign outcome、最强已支撑 claim、最大的报告风险。
+- 开始前先简洁说明：campaign outcome、选定的主指标和次级指标、最强已支撑 claim、最大的报告风险。
 - 如果某个未决前提会改变报告解释，一次只问一个问题。
 - 如果存在多种报告 framing，先给 2-3 个方案、trade-offs 和推荐项，优先最忠于证据的 framing。
 - 如果某种 framing 会实质影响后续论文 claim，要保留 approval gate。
@@ -679,27 +683,78 @@ const ZH_SKILL_FILES = {
 - 方差和显著性问题是否交代清楚？
 `,
   [path.join(".lab", ".managed", "templates", "final-report.md")]:
-`# Final Report
+`# 最终报告
 
 ## 目标
 
 - 本轮研究目标：
 - 是否达标：
 
+## 选定指标
+
+- 主指标：
+- 次级指标：
+- 必要终局证据：
+
+## 主表工件
+
+- 受管主表路径：\`<deliverables_root>/main-tables.md\`
+- 最终表现摘要：
+- 主表覆盖情况：
+
 ## 主要结果
 
-- Main finding 1:
-- Main finding 2:
+- 主要发现 1：
+- 主要发现 2：
 
 ## 失败尝试与局限
 
-- Failed attempt:
-- Limitation:
+- 失败尝试：
+- 局限：
 
 ## 仍需加强的 claims
 
-- Claim:
-- Missing support:
+- Claim：
+- 缺失支持：
+`,
+  [path.join(".lab", ".managed", "templates", "main-tables.md")]:
+`# 主表工件
+
+## 选定指标
+
+- 主指标：
+- 次级指标：
+- 必要终局证据：
+
+## 最终表现摘要
+
+- 主要结果摘要：
+- 最重要数字：
+- 报告边界：
+
+## Table 1
+
+- 作用：
+- 使用指标：
+- 最强支撑 claim：
+
+## Table 2
+
+- 作用：
+- 使用指标：
+- 最强支撑 claim：
+
+## Table 3
+
+- 作用：
+- 使用指标：
+- 最强支撑 claim：
+
+## Table 4
+
+- 作用：
+- 使用指标：
+- 最强支撑 claim：
 `,
   [path.join(".lab", ".managed", "templates", "paper-plan.md")]:
 `# 论文计划
@@ -829,7 +884,9 @@ const ZH_SKILL_FILES = {
   "results_root": "results",
   "figures_root": "figures",
   "deliverables_root": "docs/research",
-  "paper_template_root": ""
+  "paper_template_root": "",
+  "paper_template_decision": "unconfirmed",
+  "paper_template_final_reminder_acknowledged": false
 }`,
   [path.join(".lab", "context", "mission.md")]:
 `# 研究主线
@@ -995,30 +1052,33 @@ const ZH_SKILL_FILES = {
 
 ## 目标
 
-- Objective:
-- Experiment ladder:
-- Metric glossary:
-- Metric source papers:
-- Metric implementation source:
-- Comparison source papers:
-- Comparison implementation source:
-- Deviation from original implementation:
-- Terminal goal type:
-- Terminal goal target:
-- Required terminal artifact:
+- 目标:
+- 主指标:
+- 次级指标:
+- 必要终局证据:
+- 实验阶梯:
+- 指标释义:
+- 指标来源论文:
+- 指标实现来源:
+- 对比方法来源论文:
+- 对比方法实现来源:
+- 与原始实现的偏差:
+- 终止目标类型:
+- 终止目标目标值:
+- 必要终止工件:
 
 ## 结果
 
-- Status: idle
-- Goal reached: no
-- Stop reason:
-- Promotion applied: no
-- Final artifact:
-- Final rung:
-- Executed stages:
-- Iterations completed: 0
-- Started at:
-- Finished at:
+- 状态: idle
+- 目标是否达成: 否
+- 停止原因:
+- 是否已升格: 否
+- 最终工件:
+- 最终 rung:
+- 已执行阶段:
+- 已完成迭代数: 0
+- 开始时间:
+- 结束时间:
 `,
   [path.join(".lab", "context", "auto-status.md")]:
 `# 自动模式状态
@@ -1299,8 +1359,15 @@ ZH_CONTENT[path.join(".codex", "skills", "lab", "stages", "write.md")] = `# \`/l
 - 最终稿必须是 LaTeX。
 - 如果配置了 \`paper_template_root\`，先检查该模板目录，再按其结构起草论文。
 - 已接入的模板目录可能包含用户或上游修改，默认不要改模板文件。
-- 如果没有配置模板，就使用交付目录下的默认 LaTeX scaffold。
+- 如果没有配置模板且 \`paper_template_decision\` 是 \`unconfirmed\`，在第一次起草 \`.tex\` 之前必须先追问一次：继续使用默认 LaTeX scaffold，还是先接入模板目录。
+- 如果用户选择默认 scaffold，就把该决定持久化到 \`.lab/config/workflow.json\`，写成 \`paper_template_decision: default-scaffold\`。
+- 如果用户选择接入模板，就停止当前写作回合，先转到 \`superlab paper attach-template --path <dir>\`。
+- 如果 \`paper_template_decision\` 已经是 \`default-scaffold\`，就使用交付目录下的默认 LaTeX scaffold。
+- 如果当前是最终导出或最终定稿轮次、\`paper_template_root\` 仍为空、\`paper_template_decision\` 是 \`default-scaffold\`，且 \`paper_template_final_reminder_acknowledged\` 是 \`false\`，就在最终定稿前再提醒一次是否切到模板。
+- 如果用户在最终提醒里仍确认继续使用默认 scaffold，就把 \`paper_template_final_reminder_acknowledged\` 持久化为 \`true\`。
 - 只加载当前 section guide，不要一次加载全部章节参考。
+- 如果当前 section 是 \`abstract\`、\`introduction\` 或 \`method\`，还必须继续读取本地 example bank：\`references/paper-writing/examples/index.md\`、对应的 examples index，以及 1-2 个具体 example 文件。
+- 例子只能复用结构、段落角色和句法逻辑，不能直接复用原句。
 - 先写 mini-outline 再写 prose。
 - 如果缺少 framing artifact，不要继续写作，直接回到 \`/lab:framing\`。
 - 如果某个 claim 没有证据支撑，就削弱或删除。
@@ -1949,6 +2016,12 @@ ZH_CONTENT[path.join(".codex", "skills", "lab", "stages", "auto.md")] = `# \`/la
 - 先做输入归一化：把 \`Autonomy level L1/L2/L3\` 视为执行权限级别，把 \`Layer 3\`、\`Phase 1\`、\`Table 2\` 视为论文范围目标。
 - 如果用户同时提了论文层、实验 phase 和自治级别，先用一句话重述：objective、自治级别、terminal goal、scope、allowed modifications。
 - 如果 workflow language 是中文，摘要、清单条目、任务标签和进度更新都应使用中文，除非文件路径、代码标识符或字面指标名必须保持原样。
+- 当循环即将进入 \`write\`，且 \`paper_template_root\` 为空时：
+  - 如果 \`paper_template_decision\` 是 \`unconfirmed\`，必须先追问一次：继续使用默认 scaffold，还是先接入模板目录
+  - 如果用户选择默认 scaffold，就持久化 \`paper_template_decision: default-scaffold\`
+  - 如果用户选择模板，就停止当前循环并转到 \`superlab paper attach-template --path <dir>\`
+  - 如果当前是最终导出或最终定稿、\`paper_template_decision\` 是 \`default-scaffold\`，且 \`paper_template_final_reminder_acknowledged\` 是 \`false\`，就在最终定稿前再提醒一次
+  - 如果用户在最终提醒里仍确认继续使用默认 scaffold，就持久化 \`paper_template_final_reminder_acknowledged: true\`
 - 不要把 \`sleep 30\`、单次 \`pgrep\` 或一次性的 \`metrics.json\` 探针当成 rung 主命令；这些只能算进度检查。
 - 当真实实验进程还活着时，只允许发进度更新并继续等待，不能把这一 rung 当作已经完成。
 `;