superlab 0.1.13 → 0.1.14

package/README.md

@@ -187,6 +187,14 @@ superlab auto stop
 
 It does not replace manual `idea`, `data`, `framing`, or `spec` decisions.
 
+Good `/lab:auto` input is explicit. Treat `Autonomy level L1/L2/L3` as execution privilege, and treat `paper layer`, `phase`, or `table` as experiment targets. If the workflow language is Chinese, summaries, checklist items, task labels, and progress updates should also stay in Chinese unless a literal identifier must remain unchanged.
+
+Example:
+
+```text
+/lab:auto Autonomy level L2. Objective: advance paper layer 3 organizer enforcement. Terminal goal: task-completion. Scope: bounded protocol, tests, minimal implementation, and one small run. Allowed modifications: evaluator prompt registry, ingestion, and parser only.
+```
+
 ## Version
 
 Show the CLI version and the current project asset version:

package/README.zh-CN.md

@@ -185,6 +185,14 @@ superlab auto stop
 
 它不会替代手动的 `idea`、`data`、`framing`、`spec` 决策。
 
+好的 `/lab:auto` 输入应该显式写清。把 `Autonomy level L1/L2/L3` 当成执行权限级别，把 `paper layer`、`phase`、`table` 当成实验目标，不要混用。如果 workflow language 是中文，摘要、清单条目、任务标签和进度更新也应保持中文，除非某个字面标识符必须保持原样。
+
+示例：
+
+```text
+/lab:auto 自治级别 L2。目标：推进 paper layer 3 的 organizer enforcement。终止条件：完成 bounded protocol、测试、最小实现和一轮小规模结果。允许修改：evaluator prompt registry、ingestion、parser。
+```
+
 ## 版本查询
 
 查看当前 CLI 版本和当前目录项目的资产版本：

package/lib/i18n.cjs

@@ -934,12 +934,16 @@ const ZH_SKILL_FILES = {
 
 - Objective:
 - Autonomy level: L2
+- Autonomy level 只表示执行权限级别，不表示论文 layer 或 table 编号。
+- 如果你想表达论文层、实验 phase 或主表，请明确写成 \`paper layer\`、\`phase\` 或 \`table\`。
 - Approval status: draft
 - Allowed stages: run, iterate, review, report
 - Success criteria:
 - Terminal goal type:
 - Terminal goal target:
 - Required terminal artifact:
+- 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。
+- 示例 Objective: 推进 paper layer 3 的 organizer enforcement，完成一轮 bounded protocol、测试、最小实现和一轮小规模结果。
 
 ## 循环预算
 
@@ -950,6 +954,9 @@ const ZH_SKILL_FILES = {
 
 ## 阶段命令
 
+- Rung 的 \`Command\` 应该绑定真实的长任务命令，由它产出最终实验结果。
+- 短 watcher 只用于查看进度；当真实实验还在运行时，不要把短 watcher 当成 stage 或 rung 的主命令。
+- 当真实实验进程还活着时，只记录进度更新并继续等待。
 - Run command:
 - Iterate command:
 - Review command:
@@ -1383,7 +1390,7 @@ ZH_CONTENT[path.join(".lab", ".managed", "templates", "framing.md")] = `# 论文
 ZH_CONTENT[path.join(".codex", "prompts", "lab.md")] = codexPrompt(
   "查看 /lab 研究工作流总览并选择合适阶段",
   "workflow question 或 stage choice",
-  "# `/lab` for Codex\n\n`/lab` 是严格的研究工作流命令族。每次都使用同一套仓库工件和阶段边界。\n\n## 子命令\n\n- `/lab:idea`\n  调研 idea，定义问题与 failure case，归类 contribution 与 breakthrough level，对比现有方法，收束三个一眼就有意义的点，并在实现前保留 approval gate。\n\n- `/lab:data`\n  把已批准的 idea 转成数据集与 benchmark 方案，记录数据集年份、使用过该数据集的论文、下载来源、许可或访问限制，以及 classic-public、recent-strong-public、claim-specific 三类 benchmark 的纳入理由，和 canonical baselines、strong historical baselines、recent strong public methods、closest prior work 四类对比方法的纳入理由。\n\n- `/lab:auto`\n  在不改变 mission、framing 和核心 claims 的前提下，读取 eval-protocol 与 auto-mode 契约并自动编排 `run`、`iterate`、`review`、`report`，必要时扩展数据集、benchmark 和 comparison methods，并在满足升格策略时自动升级 primary package。启动前必须选定 autonomy level、声明 terminal goal，并显式批准契约。\n\n- `/lab:framing`\n  通过审计当前领域与相邻领域的术语，锁定 paper-facing 的方法名、模块名、论文题目和 contribution bullets，并在 section 起草前保留 approval gate。\n\n- `/lab:spec`\n  把已批准的 idea 转成 `.lab/changes/<change-id>/` 下的一个 lab change 目录，并在其中写出 `proposal`、`design`、`spec`、`tasks`。\n\n- `/lab:run`\n  执行最小有意义验证运行，登记 run，并生成第一版标准化评估摘要。\n\n- `/lab:iterate`\n  在冻结 mission、阈值、verification commands 与 `completion_promise` 的前提下执行有边界的实验迭代。\n\n- `/lab:review`\n  以 reviewer mode 审查文档或结果，先给短摘要，再输出 findings、fatal flaws、fix priority 和 residual risks。\n\n- `/lab:report`\n  从 runs 和 iterations 工件生成最终研究报告。\n\n- `/lab:write`\n  使用已安装 `lab` skill 下 vendored 的 paper-writing references，把稳定 report 工件转成论文 section。\n\n## 调度规则\n\n- 始终使用 `skills/lab/SKILL.md` 作为工作流合同。\n- 用户显式调用 `/lab:<stage>` 时，要立刻执行该 stage，而不是只推荐别的 `/lab` stage。\n- 先给简洁摘要，再决定是否写工件，最后回报输出路径和下一步。\n- 如果歧义会影响结论，一次只问一个问题；如果有多条可行路径，先给 2-3 个方案再收敛。\n- `/lab:spec` 前应已有经批准的数据集与 benchmark 方案。\n- `/lab:run`、`/lab:iterate`、`/lab:auto`、`/lab:report` 都应遵循 `.lab/context/eval-protocol.md`。\n- `.lab/context/eval-protocol.md` 不只定义主指标和主表，也应定义指标释义、实验阶梯，以及指标和对比实现的来源。\n- `/lab:auto` 只编排已批准边界内的执行阶段，不替代手动的 idea/data/framing/spec 决策。\n- `/lab:write` 前必须已有经批准的 `/lab:framing` 工件。\n"
+  "# `/lab` for Codex\n\n`/lab` 是严格的研究工作流命令族。每次都使用同一套仓库工件和阶段边界。\n\n## 子命令\n\n- `/lab:idea`\n  调研 idea，定义问题与 failure case，归类 contribution 与 breakthrough level，对比现有方法，收束三个一眼就有意义的点，并在实现前保留 approval gate。\n\n- `/lab:data`\n  把已批准的 idea 转成数据集与 benchmark 方案，记录数据集年份、使用过该数据集的论文、下载来源、许可或访问限制，以及 classic-public、recent-strong-public、claim-specific 三类 benchmark 的纳入理由，和 canonical baselines、strong historical baselines、recent strong public methods、closest prior work 四类对比方法的纳入理由。\n\n- `/lab:auto`\n  在不改变 mission、framing 和核心 claims 的前提下，读取 eval-protocol 与 auto-mode 契约并自动编排 `run`、`iterate`、`review`、`report`，必要时扩展数据集、benchmark 和 comparison methods，并在满足升格策略时自动升级 primary package。启动前必须选定 autonomy level、声明 terminal goal，并显式批准契约。\n\n- `/lab:framing`\n  通过审计当前领域与相邻领域的术语，锁定 paper-facing 的方法名、模块名、论文题目和 contribution bullets，并在 section 起草前保留 approval gate。\n\n- `/lab:spec`\n  把已批准的 idea 转成 `.lab/changes/<change-id>/` 下的一个 lab change 目录，并在其中写出 `proposal`、`design`、`spec`、`tasks`。\n\n- `/lab:run`\n  执行最小有意义验证运行，登记 run，并生成第一版标准化评估摘要。\n\n- `/lab:iterate`\n  在冻结 mission、阈值、verification commands 与 `completion_promise` 的前提下执行有边界的实验迭代。\n\n- `/lab:review`\n  以 reviewer mode 审查文档或结果，先给短摘要，再输出 findings、fatal flaws、fix priority 和 residual risks。\n\n- `/lab:report`\n  从 runs 和 iterations 工件生成最终研究报告。\n\n- `/lab:write`\n  使用已安装 `lab` skill 下 vendored 的 paper-writing references，把稳定 report 工件转成论文 section。\n\n## 调度规则\n\n- 始终使用 `skills/lab/SKILL.md` 作为工作流合同。\n- 用户显式调用 `/lab:<stage>` 时，要立刻执行该 stage，而不是只推荐别的 `/lab` stage。\n- 先给简洁摘要，再决定是否写工件，最后回报输出路径和下一步。\n- 如果歧义会影响结论，一次只问一个问题；如果有多条可行路径，先给 2-3 个方案再收敛。\n- `/lab:spec` 前应已有经批准的数据集与 benchmark 方案。\n- `/lab:run`、`/lab:iterate`、`/lab:auto`、`/lab:report` 都应遵循 `.lab/context/eval-protocol.md`。\n- `.lab/context/eval-protocol.md` 不只定义主指标和主表，也应定义指标释义、实验阶梯，以及指标和对比实现的来源。\n- `/lab:auto` 只编排已批准边界内的执行阶段，不替代手动的 idea/data/framing/spec 决策。\n- `/lab:write` 前必须已有经批准的 `/lab:framing` 工件。\n\n## 如何输入 `/lab:auto`\n\n- 把 `Autonomy level L1/L2/L3` 视为执行权限级别，不要和论文里的 layer、phase、table 编号混用。\n- 把 `paper layer`、`phase`、`table` 视为实验目标。例如 `paper layer 3` 或 `Phase 1 reviewer fidelity` 不是 `Autonomy level L3`。\n- 一条好的 `/lab:auto` 输入应至少说清：objective、自治级别、terminal goal、scope、allowed modifications。\n- 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。\n- 示例：`/lab:auto 自治级别 L2。目标：推进 paper layer 3 的 organizer enforcement。终止条件：完成 bounded protocol、测试、最小实现和一轮小规模结果。允许修改：evaluator prompt registry、ingestion、parser。`\n"
 );
 
 ZH_CONTENT[path.join(".codex", "prompts", "lab-data.md")] = codexPrompt(
@@ -1395,14 +1402,14 @@ ZH_CONTENT[path.join(".codex", "prompts", "lab-data.md")] = codexPrompt(
 ZH_CONTENT[path.join(".codex", "prompts", "lab-auto.md")] = codexPrompt(
   "在已批准边界内编排自动实验循环",
   "auto mode objective",
-  "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:auto`，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:auto` 阶段。它必须读取 `.lab/context/eval-protocol.md`、`.lab/context/auto-mode.md`、`.lab/context/auto-status.md` 与 `.lab/context/auto-outcome.md`，先确认 autonomy level、approval status 与 terminal goal schema，再把 eval-protocol 里的指标释义、主表计划、来源约束与结构化实验阶梯当作执行依据，在不修改 mission、framing 和核心 claims 的前提下编排已批准的 `run`、`iterate`、`review`、`report`，轮询长任务完成情况；如果声明了 rung，就保持会话活着并按 rung 转移继续推进。"
+  "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:auto`，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:auto` 阶段。它必须读取 `.lab/context/eval-protocol.md`、`.lab/context/auto-mode.md`、`.lab/context/auto-status.md` 与 `.lab/context/auto-outcome.md`，先确认 autonomy level、approval status 与 terminal goal schema，再把 eval-protocol 里的指标释义、主表计划、来源约束与结构化实验阶梯当作执行依据，在不修改 mission、framing 和核心 claims 的前提下编排已批准的 `run`、`iterate`、`review`、`report`，轮询长任务完成情况；如果声明了 rung，就保持会话活着并按 rung 转移继续推进。\n如果仓库的 workflow language 是中文，摘要、清单条目、任务标签和进度更新都必须使用中文，除非某个文件路径、代码标识符或字面指标名必须保持原样。\n把 `Layer 3`、`Phase 1`、`Table 2` 这类表达视为论文范围目标；只有显式写成 `Autonomy level L3` 或 `自治级别 L3` 时，才把它当成执行权限级别。\n不要用 `sleep 30`、单次 `pgrep` 或一次性的 `metrics.json` 探针来代替真实长任务命令；当真实实验进程还活着时，只允许发进度更新并继续等待。"
 );
 
 ZH_CONTENT[path.join(".claude", "commands", "lab.md")] = claudeCommand(
   "LAB",
   "查看 /lab 研究工作流总览并选择合适阶段",
   "workflow, research, overview",
-  "# `/lab` for Claude\n\n`/lab` 是严格的研究工作流命令族。每次都使用同一套仓库工件和阶段边界。\n\n## 子命令\n\n- `/lab:idea`\n  调研 idea，定义问题与 failure case，归类 contribution 与 breakthrough level，对比现有方法，收束三个一眼就有意义的点，并在实现前保留 approval gate。\n\n- `/lab:data`\n  把已批准的 idea 转成数据集与 benchmark 方案，记录数据集年份、使用过该数据集的论文、下载来源、许可或访问限制，以及 classic-public、recent-strong-public、claim-specific 三类 benchmark 的纳入理由，和 canonical baselines、strong historical baselines、recent strong public methods、closest prior work 四类对比方法的纳入理由。\n\n- `/lab:auto`\n  在不改变 mission、framing 和核心 claims 的前提下，读取 eval-protocol 与 auto-mode 契约并自动编排 `run`、`iterate`、`review`、`report`，必要时扩展数据集、benchmark 和 comparison methods，并在满足升格策略时自动升级 primary package。启动前必须选定 autonomy level、声明 terminal goal，并显式批准契约。\n\n- `/lab:framing`\n  通过审计当前领域与相邻领域的术语，锁定 paper-facing 的方法名、模块名、论文题目和 contribution bullets，并在 section 起草前保留 approval gate。\n\n- `/lab:spec`\n  把已批准的 idea 转成 `.lab/changes/<change-id>/` 下的一个 lab change 目录，并在其中写出 `proposal`、`design`、`spec`、`tasks`。\n\n- `/lab:run`\n  执行最小有意义验证运行，登记 run，并生成第一版标准化评估摘要。\n\n- `/lab:iterate`\n  在冻结 mission、阈值、verification commands 与 `completion_promise` 的前提下执行有边界的实验迭代。\n\n- `/lab:review`\n  以 reviewer mode 审查文档或结果，先给短摘要，再输出 findings、fatal flaws、fix priority 和 residual risks。\n\n- `/lab:report`\n  从 runs 和 iterations 工件生成最终研究报告。\n\n- `/lab:write`\n  使用已安装 `lab` skill 下 vendored 的 paper-writing references，把稳定 report 工件转成论文 section。\n\n## 调度规则\n\n- 始终使用 `skills/lab/SKILL.md` 作为工作流合同。\n- 用户显式调用 `/lab:<stage>` 时，要立刻执行该 stage，而不是只推荐别的 `/lab` stage。\n- 先给简洁摘要，再决定是否写工件，最后回报输出路径和下一步。\n- 如果歧义会影响结论，一次只问一个问题；如果有多条可行路径，先给 2-3 个方案再收敛。\n- `/lab:spec` 前应已有经批准的数据集与 benchmark 方案。\n- `/lab:run`、`/lab:iterate`、`/lab:auto`、`/lab:report` 都应遵循 `.lab/context/eval-protocol.md`。\n- `.lab/context/eval-protocol.md` 不只定义主指标和主表，也应定义指标释义、实验阶梯，以及指标和对比实现的来源。\n- `/lab:auto` 只编排已批准边界内的执行阶段，不替代手动的 idea/data/framing/spec 决策。\n- `/lab:write` 前必须已有经批准的 `/lab:framing` 工件。\n"
+  "# `/lab` for Claude\n\n`/lab` 是严格的研究工作流命令族。每次都使用同一套仓库工件和阶段边界。\n\n## 子命令\n\n- `/lab:idea`\n  调研 idea，定义问题与 failure case，归类 contribution 与 breakthrough level，对比现有方法，收束三个一眼就有意义的点，并在实现前保留 approval gate。\n\n- `/lab:data`\n  把已批准的 idea 转成数据集与 benchmark 方案，记录数据集年份、使用过该数据集的论文、下载来源、许可或访问限制，以及 classic-public、recent-strong-public、claim-specific 三类 benchmark 的纳入理由，和 canonical baselines、strong historical baselines、recent strong public methods、closest prior work 四类对比方法的纳入理由。\n\n- `/lab:auto`\n  在不改变 mission、framing 和核心 claims 的前提下，读取 eval-protocol 与 auto-mode 契约并自动编排 `run`、`iterate`、`review`、`report`，必要时扩展数据集、benchmark 和 comparison methods，并在满足升格策略时自动升级 primary package。启动前必须选定 autonomy level、声明 terminal goal，并显式批准契约。\n\n- `/lab:framing`\n  通过审计当前领域与相邻领域的术语，锁定 paper-facing 的方法名、模块名、论文题目和 contribution bullets，并在 section 起草前保留 approval gate。\n\n- `/lab:spec`\n  把已批准的 idea 转成 `.lab/changes/<change-id>/` 下的一个 lab change 目录，并在其中写出 `proposal`、`design`、`spec`、`tasks`。\n\n- `/lab:run`\n  执行最小有意义验证运行，登记 run，并生成第一版标准化评估摘要。\n\n- `/lab:iterate`\n  在冻结 mission、阈值、verification commands 与 `completion_promise` 的前提下执行有边界的实验迭代。\n\n- `/lab:review`\n  以 reviewer mode 审查文档或结果，先给短摘要，再输出 findings、fatal flaws、fix priority 和 residual risks。\n\n- `/lab:report`\n  从 runs 和 iterations 工件生成最终研究报告。\n\n- `/lab:write`\n  使用已安装 `lab` skill 下 vendored 的 paper-writing references，把稳定 report 工件转成论文 section。\n\n## 调度规则\n\n- 始终使用 `skills/lab/SKILL.md` 作为工作流合同。\n- 用户显式调用 `/lab:<stage>` 时，要立刻执行该 stage，而不是只推荐别的 `/lab` stage。\n- 先给简洁摘要，再决定是否写工件，最后回报输出路径和下一步。\n- 如果歧义会影响结论，一次只问一个问题；如果有多条可行路径，先给 2-3 个方案再收敛。\n- `/lab:spec` 前应已有经批准的数据集与 benchmark 方案。\n- `/lab:run`、`/lab:iterate`、`/lab:auto`、`/lab:report` 都应遵循 `.lab/context/eval-protocol.md`。\n- `.lab/context/eval-protocol.md` 不只定义主指标和主表，也应定义指标释义、实验阶梯，以及指标和对比实现的来源。\n- `/lab:auto` 只编排已批准边界内的执行阶段，不替代手动的 idea/data/framing/spec 决策。\n- `/lab:write` 前必须已有经批准的 `/lab:framing` 工件。\n\n## 如何输入 `/lab:auto`\n\n- 把 `Autonomy level L1/L2/L3` 视为执行权限级别，不要和论文里的 layer、phase、table 编号混用。\n- 把 `paper layer`、`phase`、`table` 视为实验目标。例如 `paper layer 3` 或 `Phase 1 reviewer fidelity` 不是 `Autonomy level L3`。\n- 一条好的 `/lab:auto` 输入应至少说清：objective、自治级别、terminal goal、scope、allowed modifications。\n- 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。\n- 示例：`/lab:auto 自治级别 L2。目标：推进 paper layer 3 的 organizer enforcement。终止条件：完成 bounded protocol、测试、最小实现和一轮小规模结果。允许修改：evaluator prompt registry、ingestion、parser。`\n"
 );
 
 ZH_CONTENT[path.join(".claude", "commands", "lab", "data.md")] = claudeCommand(
@@ -1416,7 +1423,7 @@ ZH_CONTENT[path.join(".claude", "commands", "lab", "auto.md")] = claudeCommand(
   "LAB: Auto",
   "在已批准边界内编排自动实验循环",
   "workflow, research, auto",
-  "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:auto`，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:auto` 阶段。它必须读取 `.lab/context/eval-protocol.md`、`.lab/context/auto-mode.md`、`.lab/context/auto-status.md` 与 `.lab/context/auto-outcome.md`，先确认 autonomy level、approval status 与 terminal goal schema，再把 eval-protocol 里的指标释义、主表计划、来源约束与结构化实验阶梯当作执行依据，在不修改 mission、framing 和核心 claims 的前提下编排已批准的 `run`、`iterate`、`review`、`report`，轮询长任务完成情况；如果声明了 rung，就保持会话活着并按 rung 转移继续推进。"
+  "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:auto`，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:auto` 阶段。它必须读取 `.lab/context/eval-protocol.md`、`.lab/context/auto-mode.md`、`.lab/context/auto-status.md` 与 `.lab/context/auto-outcome.md`，先确认 autonomy level、approval status 与 terminal goal schema，再把 eval-protocol 里的指标释义、主表计划、来源约束与结构化实验阶梯当作执行依据，在不修改 mission、framing 和核心 claims 的前提下编排已批准的 `run`、`iterate`、`review`、`report`，轮询长任务完成情况；如果声明了 rung，就保持会话活着并按 rung 转移继续推进。\n如果仓库的 workflow language 是中文，摘要、清单条目、任务标签和进度更新都必须使用中文，除非某个文件路径、代码标识符或字面指标名必须保持原样。\n把 `Layer 3`、`Phase 1`、`Table 2` 这类表达视为论文范围目标；只有显式写成 `Autonomy level L3` 或 `自治级别 L3` 时，才把它当成执行权限级别。\n不要用 `sleep 30`、单次 `pgrep` 或一次性的 `metrics.json` 探针来代替真实长任务命令；当真实实验进程还活着时，只允许发进度更新并继续等待。"
 );
 
 ZH_CONTENT[path.join(".codex", "skills", "lab", "SKILL.md")] = `---
@@ -1939,6 +1946,11 @@ ZH_CONTENT[path.join(".codex", "skills", "lab", "stages", "auto.md")] = `# \`/la
 - 如果契约本身不完整，一次只追问一个问题。
 - 如果存在多个可信的下一动作，先给 2-3 个 bounded 方案和推荐项，再启动长任务。
 - 只有当下一步会离开已批准的 exploration envelope、超出选定 autonomy level，或实质改变 frozen core 时，才保留人工 approval gate。
+- 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。
+- 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。
 `;
 
 ZH_CONTENT[path.join(".claude", "skills", "lab", "stages", "auto.md")] =

package/package-assets/claude/commands/lab/auto.md

@@ -9,3 +9,6 @@ Use the installed `lab` skill at `.claude/skills/lab/SKILL.md`.
 
 Execute the requested `/lab:auto` stage against the user's argument now. Do not only recommend another lab stage. If a blocking prerequisite is missing, say exactly what is missing and ask at most one clarifying question.
 This command runs the `/lab:auto` stage. It must read `.lab/context/eval-protocol.md`, `.lab/context/auto-mode.md`, `.lab/context/auto-status.md`, and `.lab/context/auto-outcome.md`, enforce the declared terminal goal schema, orchestrate approved run, iterate, review, and report stages inside that contract, poll long-running work until completion or stop conditions, and write progress plus the final outcome back into `.lab/context/auto-status.md` and `.lab/context/auto-outcome.md`.
+When the repository workflow language is Chinese, summaries, checklist items, task labels, and progress updates should be written in Chinese unless a literal identifier must stay unchanged.
+Treat `Layer 3`, `Phase 1`, or `Table 2` as paper-scope targets. Treat `Autonomy level L3` as the execution permission level.
+Do not replace the real long-running experiment command with a short watcher such as `sleep 30`, `pgrep`, or a one-shot `metrics.json` probe. While the real experiment process is still alive, emit only a progress update and keep waiting.

package/package-assets/claude/commands/lab.md

@@ -54,3 +54,17 @@ tags: [workflow, research, overview]
 - `/lab:run`, `/lab:iterate`, `/lab:auto`, and `/lab:report` should all follow `.lab/context/eval-protocol.md`, including its recorded sources for metrics and comparison implementations.
 - `/lab:write` requires an approved framing artifact from `/lab:framing`.
 - `/lab:write` requires stable report artifacts, a mini-outline, the active section guide, `paper-review.md`, and `does-my-writing-flow-source.md`, and should only change one section per round.
+
+## How to Ask for `/lab:auto`
+
+- Treat `Autonomy level L1/L2/L3` as the execution privilege level, not as a paper layer, phase, or table number.
+- Treat `paper layer`, `phase`, and `table` as experiment targets. For example, `paper layer 3` or `Phase 1 reviewer fidelity` should not be interpreted as `Autonomy level L3`.
+- A good `/lab:auto` request should name:
+  - the objective
+  - the autonomy level
+  - the terminal goal
+  - the scope or phase to advance
+  - the allowed modifications
+- If the repository workflow language is Chinese, summaries, checklist items, task labels, and progress updates should be written in Chinese unless a code identifier or file path must stay literal.
+- Good example:
+  - `/lab:auto Autonomy level L2. Objective: advance paper layer 3 organizer enforcement. Terminal goal: task-completion. Scope: bounded protocol, tests, minimal implementation, and one small run. Allowed modifications: evaluator prompt registry, ingestion, and parser only.`

package/package-assets/codex/prompts/lab-auto.md

@@ -7,3 +7,6 @@ Use the installed `lab` skill at `.codex/skills/lab/SKILL.md`.
 
 Execute the requested `/lab:auto` stage against the user's argument now. Do not only recommend another lab stage. If a blocking prerequisite is missing, say exactly what is missing and ask at most one clarifying question.
 This command runs the `/lab:auto` stage. It must read `.lab/context/eval-protocol.md`, `.lab/context/auto-mode.md`, `.lab/context/auto-status.md`, and `.lab/context/auto-outcome.md`, enforce the declared terminal goal schema, orchestrate approved run, iterate, review, and report stages inside that contract, poll long-running work until completion or stop conditions, and write progress plus the final outcome back into `.lab/context/auto-status.md` and `.lab/context/auto-outcome.md`.
+When the repository workflow language is Chinese, summaries, checklist items, task labels, and progress updates should be written in Chinese unless a literal identifier must stay unchanged.
+Treat `Layer 3`, `Phase 1`, or `Table 2` as paper-scope targets. Treat `Autonomy level L3` as the execution permission level.
+Do not replace the real long-running experiment command with a short watcher such as `sleep 30`, `pgrep`, or a one-shot `metrics.json` probe. While the real experiment process is still alive, emit only a progress update and keep waiting.

package/package-assets/codex/prompts/lab.md

@@ -52,3 +52,17 @@ argument-hint: workflow question or stage choice
 - `/lab:run`, `/lab:iterate`, `/lab:auto`, and `/lab:report` should all follow `.lab/context/eval-protocol.md`, including its recorded sources for metrics and comparison implementations.
 - `/lab:write` requires an approved framing artifact from `/lab:framing`.
 - `/lab:write` requires stable report artifacts, a mini-outline, the active section guide, `paper-review.md`, and `does-my-writing-flow-source.md`, and should only change one section per round.
+
+## How to Ask for `/lab:auto`
+
+- Treat `Autonomy level L1/L2/L3` as the execution privilege level, not as a paper layer, phase, or table number.
+- Treat `paper layer`, `phase`, and `table` as experiment targets. For example, `paper layer 3` or `Phase 1 reviewer fidelity` should not be interpreted as `Autonomy level L3`.
+- A good `/lab:auto` request should name:
+  - the objective
+  - the autonomy level
+  - the terminal goal
+  - the scope or phase to advance
+  - the allowed modifications
+- If the repository workflow language is Chinese, summaries, checklist items, task labels, and progress updates should be written in Chinese unless a code identifier or file path must stay literal.
+- Good example:
+  - `/lab:auto Autonomy level L2. Objective: advance paper layer 3 organizer enforcement. Terminal goal: task-completion. Scope: bounded protocol, tests, minimal implementation, and one small run. Allowed modifications: evaluator prompt registry, ingestion, and parser only.`

package/package-assets/shared/lab/context/auto-mode.md

@@ -6,14 +6,18 @@ If `eval-protocol.md` declares structured rung entries, auto mode follows those
 
 ## Objective
 
-- Objective:
+- Objective: 
 - Autonomy level: L2
+- Autonomy level controls execution privilege, not paper layer or table number.
+- If you mean a paper layer, phase, or table, spell it explicitly as `paper layer`, `phase`, or `table`.
 - Approval status: draft
 - Allowed stages: run, iterate, review, report
 - Success criteria:
 - Terminal goal type:
 - Terminal goal target:
 - Required terminal artifact:
+- If the workflow language is Chinese, keep summaries, checklist items, task labels, and progress updates in Chinese.
+- Example objective: advance paper layer 3 organizer enforcement through one bounded protocol, tests, minimal implementation, and one small run.
 
 ## Loop Budget
 
@@ -24,6 +28,9 @@ If `eval-protocol.md` declares structured rung entries, auto mode follows those
 
 ## Stage Commands
 
+- Rung `Command` should be the real long-running command that owns the experiment result.
+- A short watcher is only a progress probe. Do not use a short watcher as the stage or rung command when the real experiment is still running.
+- While the real experiment process is still alive, only record a progress update and keep waiting.
 - Run command:
 - Iterate command:
 - Review command:

package/package-assets/shared/skills/lab/stages/auto.md

@@ -54,6 +54,8 @@
 - You may promote exploratory additions to the primary package only when the contract's promotion policy is satisfied and the promotion is written back into `data-decisions.md`, `decisions.md`, `state.md`, and `session-brief.md`.
 - Poll long-running commands until they finish, hit a timeout, or hit a stop condition.
 - Keep a poll-based waiting loop instead of sleeping blindly.
+- Do not treat a short watcher such as `sleep 30`, a one-shot `pgrep`, or a single `metrics.json` probe as the rung command when the real experiment is still running.
+- Bind each rung to the real long-running command or process that owns the experiment result.
 - Always write a canonical `.lab/context/auto-outcome.md` when the run completes, stops, or fails.
 - When the evaluation protocol declares structured ladder rungs, execute them as a foreground rung state machine:
   - each rung must declare `Stage`, `Goal`, `Command`, `Watch`, `Gate`, `On pass`, `On fail`, and `On stop`
@@ -87,3 +89,21 @@
 - If the contract is incomplete, ask one clarifying question at a time.
 - If multiple next actions are credible, present 2-3 bounded options with trade-offs before arming a long run.
 - Only ask for approval when the next step would leave the approved exploration envelope, exceed the chosen autonomy level, or materially change the frozen core.
+
+## Input Normalization
+
+- Normalize ambiguous user requests before arming the loop.
+- Treat `Autonomy level L1/L2/L3` as execution privilege only.
+- Treat `Layer`, `Phase`, and `Table` references as paper-structure or experiment-scope targets, not as autonomy levels.
+- Example:
+  - `Layer 3 organizer enforcement` means a paper layer or experiment target.
+  - `Autonomy level L3` means the aggressive campaign permission envelope.
+- If the user mixes framework work and experiment work in one request, restate a normalized contract with:
+  - objective
+  - autonomy level
+  - terminal goal
+  - scope
+  - allowed modifications
+- Then ask at most one clarifying question if a blocking field is still missing.
+- If `.lab/config/workflow.json` sets the workflow language to Chinese, write summaries, options, checklist items, task labels, and progress updates in Chinese unless a file path, code identifier, or literal metric name must remain unchanged.
+- While the real experiment process is still alive, emit only a progress update and keep waiting. Do not present a terminal summary for that rung until the process exits or the rung hits an explicit stop boundary.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "superlab",
-  "version": "0.1.13",
+  "version": "0.1.14",
   "description": "Strict /lab research workflow installer for Codex and Claude",
   "keywords": [
     "codex",