agentworkshop 0.7.49 → 0.7.50

package/.output/server/index.mjs

@@ -15,17 +15,17 @@ import 'url';
 import 'node:crypto';
 import 'node:events';
 import 'node:buffer';
-import 'node:url';
 import 'node:fs';
 import 'node:path';
-import 'node:fs/promises';
-import 'node:os';
+import 'node:url';
 import 'js-yaml';
+import 'node:os';
+import 'zod';
 import 'node:child_process';
-import 'node:net';
-import 'node:util';
 import 'node:sqlite';
 import 'node:module';
+import 'node:fs/promises';
+import 'node:net';
+import 'node:util';
 import 'node:zlib';
-import 'zod';
 //# sourceMappingURL=index.mjs.map

package/.output/server/package.json

@@ -1,6 +1,6 @@
 {
   "name": "agentworkshop-prod",
-  "version": "0.7.49",
+  "version": "0.7.50",
   "type": "module",
   "private": true,
   "dependencies": {

package/README-zh.md

@@ -28,7 +28,7 @@
 
 **[English](./README.md)** · **[在线文档](https://kingdol666.github.io/AgentWorkShop)** · **[版本发布](https://github.com/kingdol666/AgentWorkShop/releases)** · **[更新日志](./changelog.md)** · **[插件 API](./docs/plugins.md)** · **[SDK](./docs/sdk.md)**
 
-<sub><b>v0.7.49</b> · 14 个执行引擎 · 6 种现场协议（5 内置 + 串口插件） · 111 个运行时设置项 · 双语文档（简体中文 / English）</sub>
+<sub><b>v0.7.50</b> · 14 个执行引擎 · 6 种现场协议（5 内置 + 串口插件） · 113 个运行时设置项 · 双语文档（简体中文 / English）</sub>
 
 <br />
 
@@ -79,7 +79,7 @@
 7 状态任务机 · 根任务队列 + 执行租约<br/>
 群聊作业 · 定时任务 · HITL 审批<br/>
 FTS5 + 向量记忆 · 插件热重载<br/>
-SDK · CLI · TUI · 约 1100+ 条验收断言
+SDK · CLI · TUI · 外部 MCP · 约 1100+ 条验收断言
 
 </td>
 </tr>
@@ -96,7 +96,7 @@ SDK · CLI · TUI · 约 1100+ 条验收断言
 | **上手指南** | [指南](https://kingdol666.github.io/AgentWorkShop/guide/getting-started) | `docs/site/guide/` | 安装 → 首次运行 → 第一次「Agent × 产线」会话、配置系统、数采协议、数控读写、HITL、配方版本、多 Harness 团队、产线权限、AML |
 | **插件开发** | [插件指南](https://kingdol666.github.io/AgentWorkShop/plugins/) | [`docs/plugins.md`](./docs/plugins.md) | 完整扩展契约：三种作用域、`index.mjs` 清单、`ctx` 服务端/浏览器能力面、`settings`/`groups` 声明、生命周期事件、i18n、面板、团队级开关、一个真实案例 |
 | **SDK** | [SDK 指南](https://kingdol666.github.io/AgentWorkShop/sdk/) | [`docs/sdk.md`](./docs/sdk.md) | `agentworkshop/sdk` 作为编程客户端（带信封处理的平台 REST 客户端）与插件扩展基座；TypeScript 类型；浏览器端 SDK |
-| **CLI** | [CLI 手册](https://kingdol666.github.io/AgentWorkShop/cli/) | [`docs/cli.md`](./docs/cli.md) | 全部 14 个 `aw` 指令、全局选项、退出码、双模式路径模型、指令注册系统、维护者发布指引 |
+| **CLI** | [CLI 手册](https://kingdol666.github.io/AgentWorkShop/cli/) | [`docs/cli.md`](./docs/cli.md) | 全部 15 个 `aw` 指令、全局选项、退出码、双模式路径模型、指令注册系统、维护者发布指引 |
 | **AML** | [AML 指南](https://kingdol666.github.io/AgentWorkShop/guide/aml) | [`docs/aml.md`](./docs/aml.md) | 自动建模实验室：数据集构建、作业编排、模型注册表、晋级门禁、Agent 工具 |
 | **TUI** | — | [`docs/tui.md`](./docs/tui.md) · [`tui/README.md`](./tui/README.md) | 终端工作台：频道、Agent、任务、实时监控、HITL 应答 |
 | **多 Harness 架构** | — | [`docs/multi-harness-architecture.md`](./docs/multi-harness-architecture.md) | 引擎分类、归一化会话契约、供应商/模型目录、失败模式 |
@@ -178,10 +178,14 @@ AgentWorkShop 起家于**多智能体软件工作坊**——Channel 内的编码
 | **全操作审计日志** | 用户 / Agent / 系统 的每个动作都落进同一份可检索日志；操作者归属「Channel名/成员名」，与用户和系统天然区分。经 WS 实时推送。 |
 | **团队级插件开关** | 每个团队（Channel）持有**独立插件开关组**（`channel_plugins`）：被关闭插件的工具不注入该团队 Agent。插件本体经 `aw plugin` 与 `/plugins` 页热管理。 |
 | **插件扩展 API** | `plugins/<name>/` 下的一个自包含目录**同时增强两半**：`index.mjs`（服务端：钩子、路由、Agent 工具、**数采读驱动 / DCW 写驱动**/帧处理器/节点模板、配置分组、KV、定时器）与 `client.mjs`（浏览器：注入具名插槽的面板、i18n、设置 UI）。三种作用域——`builtin`（随包发布）> `project`（检出）> `user`（`~/.AgentWorkShop`）——启停**与代码修改**均有约 1 秒热重载；停用插件的驱动随热重载立即摘除。内置示例 **serial-bridge**（串口通信：读/写驱动 + 串口探针 API + 前端面板）。完整契约见 [`docs/plugins.md`](./docs/plugins.md)。 |
-| **AML —— 自动建模实验室** | 数据集构建 → 训练作业 → 排行榜 → 晋级门禁 → 模型引用，全部可在 `/aml` 页驱动，也可由 Agent 通过 10 个 `aml_*` 工具驱动。Python 运行时由 `uv` 引导至 `./aml` 资产根；产物与元数据都留在配置根下。 |
-| **AML 混合孪生 × MPC** | 灰箱物理主干 + 有界 PyTorch 残差；`TwinSnapshot` / `VirtualTrial` / `RecommendationCertificate` 配 UQ/OOD 筛查与全轨迹门禁；`hybrid_twin` profile 会额外注入 6 个工具（`twin_*`、`mpc_optimize`）。试验按设计保持虚拟（`candidateExecuted=false`），证书签发前只做推荐、不落真实写入。 |
+| **AML —— 自动建模实验室** | 数据集构建 → 训练作业 → 排行榜 → 晋级门禁 → 模型引用，全部可在 `/aml` 页驱动，也可由 Agent 通过 **14 个 `aml_*` 工具**驱动。Python 运行时由 `uv` 引导至配置根内的 `aml/` 资产根；产物与元数据都留在配置根下。模型带人读身份——label（产线 · 配方 · 目标）、description 与产线/目标 id——可用 `aml_model_find` 按场景选型。 |
+| **AML 混合孪生 × MPC** | 灰箱物理主干 + 有界 PyTorch 残差；`TwinSnapshot` / `VirtualTrial` / `RecommendationCertificate` 配 UQ/OOD 筛查与全轨迹门禁；`hybrid_twin` profile 会额外注入 **12 个 profile 工具**（场景发现/编译/冻结、物理 spec 起草/校验/编译、快照、试验、MPC、门禁、Provider 目录）与通用工具 `twin_calibration_request`。试验按设计保持虚拟（`candidateExecuted=false`），证书签发前只做推荐、不落真实写入。 |
+| **双模式优化闭环** | 专属频道模板把闭环**结构性拆开**：*aml-training* 频道拿建模面（数据集、作业、场景编译、物理 spec、建模任务——无产线写入、无 MPC），*aml-optimization* 频道拿执行面（快照、试验、门禁、MPC、贝叶斯寻优 `twin_bayes_optimize`、受治理微写 `optimization_explore`）——边训边优在结构上不可能。**探索**模式走治理链写入并做响应回读配对（每笔写入落 `optimization_explorations` 台账）；**生产**模式要求绑定过门禁的模型（`aml` 模式，实例化时给 `boundModelId`）。 |
 | **运行时可观测** | `GET /api/system/monitor` 把 Agent 团队内部暴露成数字：根队列深度、看门狗介入次数、Harness 会话复用、记忆 outbox 积压——每项都有对应的文档化回退开关，新机制可以不重新部署就关掉。 |
-| **全量配置驱动运行时** | 全部运行旋钮（记忆预算、上下文压缩、回退护栏、保留策略、备份、日志级别…）在设置描述符注册表声明一次，优先级 **config.yml < runtime-settings < env**——**111 个设置项、16 组**（32 live / 79 restart），代码零硬编码默认。 |
+| **产线操作单页（`/operations`）** | 一页看全对产线的每一次触碰：实时操作流水（REST 快照 + WS 推流）把写控、数采、建模、配方、回退、报警、产线、人工、系统动作归入**九类分类**，写控行渲染结构化 **前值 → 后值 diff 徽章**（失败标红），并直接从事件流推断**闭环阶段**（监视 → 探索·小步 → 推荐搜索 → 影子验证 → 生产）。同一数据面也供工作台控制台的「产线作业」面板使用。 |
+| **外部 MCP 网关** | `aw mcp` 启动零依赖的 **stdio MCP 服务**（38 个工具），自动发现运行中的实例（环境变量 → 锁文件 → 配置文件），把产线、配方（含 mark-good）、受治理 DCW 写入、工艺参数语义映射、数采查询、频道、团队供给、模型晋升、判卷/回退与插件管理暴露给任意 MCP 客户端——受 `mcp.enabled` 运行时设置门控，附 `--doctor` 与 `--print-config` 辅助指令。 |
+| **全量配置驱动运行时** | 全部运行旋钮（记忆预算、上下文压缩、回退护栏、保留策略、备份、日志级别、系统时区、MCP 开关…）在设置描述符注册表声明一次，优先级 **config.yml < runtime-settings < env**——**113 个设置项、18 组**（34 live / 79 restart），代码零硬编码默认。 |
+| **可配置系统时区** | 一个 `time.timeZone` 设置项（IANA 时区，live 热生效）统一驱动人读时间——日志、聊天、审计、Agent 记录与定时任务；机器时间轴（数采样本、`timestamptz`）仍保存绝对时间。 |
 | **可配置节拍** | 采样与查询的默认值/下限全部是 **live 设置**（`daq.sampling.*`、`daq.query.*`）：热重载、create/patch 时钳制，Agent 工具描述实时携带当前值。 |
 
 #### 数字孪生
@@ -258,7 +262,7 @@ AgentWorkShop 起家于**多智能体软件工作坊**——Channel 内的编码
 <td width="50%"><img src="https://raw.githubusercontent.com/kingdol666/AgentWorkShop/main/docs/readme-assets/shot-monitor.png" alt="运行时监控" width="100%" /><br/><sub><b>运行时监控。</b>每条已接线频道、成员数、依赖环与归属用户。</sub></td>
 </tr>
 <tr>
-<td><img src="https://raw.githubusercontent.com/kingdol666/AgentWorkShop/main/docs/readme-assets/shot-settings.png" alt="系统设置" width="100%" /><br/><sub><b>系统设置。</b>16 组 111 个设置项，由描述符驱动——CLI 读的是同一份注册表。</sub></td>
+<td><img src="https://raw.githubusercontent.com/kingdol666/AgentWorkShop/main/docs/readme-assets/shot-settings.png" alt="系统设置" width="100%" /><br/><sub><b>系统设置。</b>18 组 113 个设置项，由描述符驱动——CLI 读的是同一份注册表。</sub></td>
 <td><img src="https://raw.githubusercontent.com/kingdol666/AgentWorkShop/main/docs/readme-assets/shot-plugins.png" alt="插件管理" width="100%" /><br/><sub><b>插件管理。</b>三种作用域、改代码即热重载、按团队开关。</sub></td>
 </tr>
 </table>
@@ -382,7 +386,7 @@ aw update --check                      # 只报告，不安装
 npm install -g agentworkshop@latest    # 手动等效
 ```
 
-版本遵循 semver。每次 `aw start` 都会校验配置根，并把 `home` 之前的旧版 `data/` 布局迁移进来（以最新文件为准），因此数据可以跨版本存活。SQLite schema 迁移在服务端启动时执行。当前版本：**v0.7.49**——见[版本发布](https://github.com/kingdol666/AgentWorkShop/releases)。
+版本遵循 semver。每次 `aw start` 都会校验配置根，并把 `home` 之前的旧版 `data/` 布局迁移进来（以最新文件为准），因此数据可以跨版本存活。SQLite schema 迁移在服务端启动时执行。当前版本：**v0.7.50**——见[版本发布](https://github.com/kingdol666/AgentWorkShop/releases)。
 
 ### 第一次「Agent × 产线」会话（约 2 分钟）
 
@@ -431,6 +435,7 @@ aw config set daq.query.minBucketMs 500             # 查询下限（samples/产
 | `aw stop` | 依单实例锁终止运行中的 aw 服务实例 |
 | `aw config list · get · set · unset · reset` | 读写运行时设置（schema 校验 + 原子写盘） |
 | `aw plugin list · create · enable · disable` | 跨三种作用域管理插件：查看、脚手架（默认 project 作用域；`--global` 为 user 作用域）、启停（写状态文件，运行中的服务约 1 秒热重载） |
+| `aw mcp` | 工业 MCP stdio 服务（38 个工具，覆盖平台 REST：产线、配方、DCW 写入、参数映射、数采、频道、团队、模型晋升、插件）——自动发现运行实例；`--doctor` 一次讲清发现/健康/鉴权，`--print-config` 直出可粘贴的 MCP 客户端配置 |
 | `aw home` | 查看/初始化配置根 `.AgentWorkShop` |
 | `aw init <dir>` | 脚手架一个可运行的新项目（含完整配置系统与 CLI） |
 | `aw register <路径\|URL\|npm:包名>` | 注册一条新指令——项目级或 `--global` 用户级 |
@@ -491,10 +496,14 @@ export async function run(argv, ctx) {
 
 这是闭环里的建模半边：`/aml` 用已打标的遥测构建**数据集**，把**训练作业**提交到 `uv` 托管的 Python 运行时，在**排行榜**上给各次运行排名，并且只有跨过配置的**门禁**（NRMSE、rollout NRMSE、验证/测试差距、最小行数与运行次数）才允许晋级。晋级后的模型按 id 引用，因此 MPC 或影子孪生控制器消费的是有版本号的产物，而不是含糊的「最新」。
 
-- Agent 通过 10 个工具做同样的事（`aml_dataset_build`、`aml_job_submit`、`aml_job_status`、`aml_leaderboard`、`aml_model_promote`……）——提交一个目标，让团队去训练并汇报。
+- Agent 通过 14 个工具做同样的事（`aml_dataset_build`、`aml_job_submit`、`aml_job_status`、`aml_leaderboard`、`aml_model_promote`、`aml_model_find`、`aml_training_plan_*`……）——提交一个目标，让团队去训练并汇报。
 - Python 运行时由 `uv` 自行引导进配置根内的 `./aml` 资产根；数据集/产物/元数据路径、磁盘配额、作业超时、并发与保留策略全部是**设置项**（`aml` 组共 16 项），而不是常量。
 - 治理默认保守：跨配方数据集除非显式允许否则拒绝，每个作业都有归属。
-- **混合孪生 × MPC 内核**：AML 同时承载**混合孪生**平面——`SceneContract` / `PhysicsModelManifest` / `TwinSnapshot` / `ObjectiveProfile` / `VirtualTrial` / `RecommendationCertificate`，底座是带**有界 PyTorch 残差**的灰箱物理主干，外加 UQ/OOD 筛查与全轨迹硬约束门禁。频道通过 `hybrid_twin` profile 显式启用，启用后会注入 6 个专属工具（`twin_scene_read`、`twin_snapshot_create`、`twin_trial_run`、`mpc_optimize`、`twin_gate_evaluate`、`twin_calibration_request`）；试验按设计就是**虚拟**的（`candidateExecuted=false`），只有门禁与收益同时通过才签发推荐证书。服务端策略两档：`safe_small_step`(数据不足) 与 `precise_search`(模型过门禁)。计划与验收记录见 [`docs/aml-hybrid-twin-mpc-integration-plan.md`](./docs/aml-hybrid-twin-mpc-integration-plan.md)、[`docs/aml-hybrid-twin-implementation-acceptance.md`](./docs/aml-hybrid-twin-implementation-acceptance.md)。
+- **混合孪生 × MPC 内核**：AML 同时承载**混合孪生**平面——`SceneContract` / `PhysicsModelManifest` / `TwinSnapshot` / `ObjectiveProfile` / `VirtualTrial` / `RecommendationCertificate`，底座是带**有界 PyTorch 残差**的灰箱物理主干，外加 UQ/OOD 筛查与全轨迹硬约束门禁。频道通过 `hybrid_twin` profile 显式启用，启用后会注入 12 个 profile 专属工具（`twin_provider_catalog`、`twin_scene_discover/compile/freeze`、`twin_physics_spec_draft/validate/compile`、`twin_scene_read`、`twin_snapshot_create`、`twin_trial_run`、`mpc_optimize`、`twin_gate_evaluate`）与通用工具 `twin_calibration_request`；试验按设计就是**虚拟**的（`candidateExecuted=false`），只有门禁与收益同时通过才签发推荐证书。服务端策略两档：`safe_small_step`(数据不足) 与 `precise_search`(模型过门禁)。计划与验收记录见 [`docs/aml-hybrid-twin-mpc-integration-plan.md`](./docs/aml-hybrid-twin-mpc-integration-plan.md)、[`docs/aml-hybrid-twin-implementation-acceptance.md`](./docs/aml-hybrid-twin-implementation-acceptance.md)。
+- **任意场景建模闭环**：闭环不绑定内置场景。绑定节点 → `twin_scene_discover` 带证据推断 control/state/disturbance/target/guard 语义 → `twin_scene_compile` 生成 `SceneContract` 草稿 → 您确认 → `twin_scene_freeze` 冻结（版本不可覆盖）→ `twin_physics_spec_draft` 起草骨架 PhysicsSpec（每个状态一条一阶弛豫方程，τ/增益为带先验盒的可校准参数）供您润色 → 校验/编译（白名单 AST，禁任意代码）→ `aml_job_submit { job_kind: 'hybrid_residual' }` **可不带代码**：内置参考训练器先做物理参数校准（stage A）、再拟合 3 成员有界残差集成（stage B）、最后报告 conformal q90 覆盖率（stage D）——校准后的 spec 回写，评估/预测/试验三侧同源。
+- **模型驱动孪生与持续校准**：`twin_trial_run` / `mpc_optimize` 可传 `model_id`——rollout 由训练出的 hybrid 模型驱动（校准后物理主干 + 残差集成 ONNX，`ModelBackedHybridProvider`），UQ 覆盖率与成员分歧来自训练工件；模型数据集与场景的产线/产品/配方不一致直接拒绝（`TWIN_MODEL_SCENE_MISMATCH`）。持续校准请求由 60s 定时 worker 消费，自动提交 `hybrid_residual` 重训**候选**；晋升仍走 HITL + Twin Gate（`AML_TWIN_WORKER_DISABLED=1` 可整体关闭）。
+- **双模式优化闭环**：随包附带两个频道模板——`chtpl-aml-training-default`（建模面：22 个工具，无产线写入、无 MPC）与 `chtpl-aml-optimization-default`（执行面：快照/试验/门禁/MPC/贝叶斯寻优，外加 `optimization_explore`——受治理微写，每笔设定值与响应回读配对落 `optimization_explorations` 台账）。两面按频道互斥注入；优化频道实例化时可选**探索**模式（治理小步，无需模型）或 **aml** 模式（必须给 `boundModelId`——过门禁的模型先绑定，执行面才解锁）。
+- **孪生 Provider 插件 SDK（热加载）**：工业场景的物理主干 / 场景包 / 目标 / 求解器经 `ctx.twin` 注册进 **Twin Provider Registry**（`registerPhysicsProvider` / `registerScenePack` / `registerObjectiveProfile` / `registerTrainingAdapter` / `registerSolverAdapter`，另有 `listProviders` / `getProviderHealth` / `resolveProvider` / `validateProvider` / `retireProvider` / `isRegistryAvailable`）。孪生工具按 `SceneContract.physicsProfileId` 解析 Provider（解析不到回退内置 `twin-injection-default`），**换场景不改工具与频道**；热重载按 **generation** 装载新代、排空旧代，试验与 MPC 持租约 ⇒ 在飞工作不被作废，`retireProvider` 等引用归零才下线。系统自带注塑 Provider 走的正是这套外部契约（`server/plugins-builtin/twin-injection-default/`），装法与其它插件同址、停用或热重载无需重启。清单字段与可运行示例见 [`docs/plugins.md`](./docs/plugins.md)「混合孪生 Provider 插件」，SDK 面见 [`docs/sdk.md`](./docs/sdk.md) §4.10。
 
 ---
 
@@ -596,12 +605,12 @@ SUBMITTED ─▶ ASSIGNED ─▶ WORKING ─▶ WAITING ─▶ COMPLETED
 | 真引擎协作 | 任务流真跑 **PASS**、`e2e-omp-workspace` **13 / 13**、群聊全真 **PASS**、多用户 HIL **PASS**（2026-09-24） | 真实 `omp` lead + worker：分解/派发/进度/成果/workspace 文件读取与卸载；多用户 HITL 可见性；多引擎团队 14/18（omp 与 dsh 全绿） | `npx tsx scripts/e2e-agentteam-task-flow-real.ts` · `npx tsx scripts/e2e-omp-workspace.ts` |
 | 前端验收 | 响应式门禁 **16 / 16** 页、抽屉交互 **10 / 10**（2026-09-24，真实 Chromium） | 全部产品页（含 3D 小镇）在门禁阈值下渲染通过；抽屉开合、滚动锁、路由跳转、零 pageerror | `node scripts/ui/verify-responsive.mjs --routes all` · `node scripts/ui/verify-nav.mjs` |
 | `_aw0746-stage12.mjs` | **10 / 0**（2026-09-25,全局安装的 **v0.7.49 打包系统**） | 健康门(version=0.7.49)→ 管理员登录/角色面 → 创建 Channel(真实 `omp` lead)→ 开群聊 → **真实 omp 群聊回复** → 系统监控可见 omp 子进程 | `AW_BASE=<base> node scripts/_aw0746-stage12.mjs` |
-| `_aw0746-stage34.mjs` | **11 / 0**（2026-09-25,v0.7.49 打包系统 + PLC 模拟器） | 模拟器建线（11 数控 + 14 数采）→ 产品/配方三元组 → 开跑后**样本逐条带 recipe_id/run_id** → `daq_query` **默认只取当前活动批次**（`scope=all` 可放开，实测 1 vs 4）→ 实例化 Hybrid Twin 模板（lead + 3 worker,注入孪生 6 件套 / 总 57 工具） | `AW_BASE=<base> node scripts/_aw0746-stage34.mjs` |
+| `_aw0746-stage34.mjs` | **13 / 0**（2026-09-25,v0.7.49 打包系统 + PLC 模拟器） | 模拟器建线（11 数控 + 14 数采）→ 产品/配方三元组 → 开跑后**样本逐条带 recipe_id/run_id** → `daq_query` **默认只取当前活动批次**（`scope=all` 可放开，实测 1 vs 4）→ 实例化 Hybrid Twin 模板（lead + 3 worker,注入孪生 6 件套 / 总 57 工具） | `AW_BASE=<base> node scripts/_aw0746-stage34.mjs` |
 | `_aw0746-stage6b-train-sweep.mjs` | 门禁**全绿**（2026-09-25,v0.7.46 打包系统,真实 torch 训练） | 1 s 拍格数据集（**1724 行 / 10 批次**）→ 内联 code 作业真实训练 → G1 单步 **0.0992**（≤0.10）/ G2 滚动 0.1092（≤0.25）/ G3 泛化 14.6%（≤20%）/ G4 行数与批次 → 模型注册并**晋升 production** | `AW_BASE=<base> AW_DATASET_ID=<ds> node scripts/_aw0746-stage6b-train-sweep.mjs` |
 | `_aw0746-stage7-twin.mjs` | **17 / 0**（2026-09-25,v0.7.49 打包系统,真实 DAQ + 真实模型） | 真实样本 → TwinSnapshot `fresh=true` → 65 组真实 VirtualTrial（12 条安全候选,`candidateExecuted=false`）→ 12 项推荐门禁 **write_eligible** → 门禁结论回写模型 `twinEligibility` → `mpc_optimize` 升档 **precise_search** → 全程 **0 DCW 写入** | `AW_BASE=<base> AW_REQUIRE_MODEL=1 node scripts/_aw0746-stage7-twin.mjs` |
 | `_aw0746-stage6c-jobsmoke.mjs` | **5 / 0**（2026-09-25,v0.7.49 打包系统） | AML 作业链路：内联 `code` 被接受（不再要求 `workspace/train.py`）→ 作业到终态 → 训练日志/ONNX 痕迹 → 实验与逐项门禁落库 | `AW_BASE=<base> node scripts/_aw0746-stage6c-jobsmoke.mjs` |
 | `_aw0748-failclosed-probe.mjs` | **7 / 0**（2026-09-25,v0.7.49 打包系统） | 硬约束**失败关闭**回归:auto_daq 按真实绑定取样建快照 → 可映射约束 `weight` 正常评估 → 不可映射的 `part_weight` 判不通过并给出明细（v0.7.47 会假通过）→ 空轨迹同样判不通过 | `AW_BASE=<base> node scripts/_aw0748-failclosed-probe.mjs` |
-| `_aw0748-mock-closedloop.mjs` | **11 / 0**（2026-09-25,v0.7.49 打包系统,进程内 `mock`,零额度） | 调度闭环回归:创建 mock Channel → 根任务被真实派发并收口 → `[mock:complex]` 强制委派出子任务 → 父任务汇总收口 → **第二个/第三个根任务同样被派发**（修复前:lead 空闲卸载后 `scheduler` 置空,后续根任务只落库、一直 SUBMITTED 直到 `ROOT_TIMEOUT`）→ mock 无 host tool 通道时工具桥显式拒绝 | `AW_BASE=<base> node scripts/_aw0748-mock-closedloop.mjs` |
+| `_aw0748-mock-closedloop.mjs` | **13 / 0**（2026-09-25,v0.7.49 打包系统,进程内 `mock`,零额度） | 调度闭环回归:创建 mock Channel → 根任务被真实派发并收口 → `[mock:complex]` 强制委派出子任务 → 父任务汇总收口 → mock worker 就位后 [mock:complex] 委派子任务并汇总 → **第二个/第三个根任务同样被派发**（修复前:lead 空闲卸载后 `scheduler` 置空,后续根任务只落库、一直 SUBMITTED 直到 `ROOT_TIMEOUT`）→ mock 无 host tool 通道时工具桥显式拒绝 | `AW_BASE=<base> node scripts/_aw0748-mock-closedloop.mjs` |
 | 工业基准——integrated | **83 / 83 检查，硬门禁绿**（2026-09-21） | 五协议五产线基准：连通性、采样入 Timescale、受控下发、治理、Agent 闭环、工艺参数语义层 | `node bench/pipeline.mjs --profile integrated` |
 | 多场景闭环 | 4 场景 · 闭环达标（2026-09-21） | 注塑克重窗口寻优 · A²/O 污水排放达标 + 能耗最小化 · 连续退火质量窗与产能权衡 · BOPET 产线任务 | `node bench/scenarios.mjs --scenarios injection,wwtp,anneal` |
 | 产线权限 / 审计负向 | 21 / 21 + 9 / 9（2026-09-12） | 三态产线授权 + 人类可读 403、绑定主体校验、授权撤销收敛、无 token WS 零遥测 | `node scripts/_dbg-perms-e2e.mjs <base> <adminPass>` · `node scripts/_dbg-audit-neg-e2e.mjs <base> <adminPass>` |
@@ -622,7 +631,7 @@ AgentWorkShop/
 │   ├── pages/                  # / · /workshop · /workshop/agents · /workshop/teams
 │   │                           # /workshop/channel-templates · /workshop/schedules · /workshop/w/:id
 │   │                           # /town · /daq · /daq/:id · /dcw · /dcw/:id
-│   │                           # /aml · /monitor · /logs · /permissions
+│   │                           # /aml · /operations · /monitor · /logs · /permissions
 │   │                           # /plugins · /users · /tokens · /settings
 │   ├── components/workshop/    # 时间线 · 泳道 · 任务板 · 记忆面板 · 3D 小镇
 │   └── stores/composables/     # Pinia + AEP 客户端
@@ -633,13 +642,14 @@ AgentWorkShop/
 │   │   ├── agents/             # AgentInterface：14 个引擎（+ 工业工具）
 │   │   ├── daq/ dcw/ aml/      # 边缘运行时 · 驱动 · 队列 · 存储 · 建模实验室
 │   │   └── db/                 # node:sqlite 仓储层
-│   ├── mcp/                    # MCP 服务（25 个工具）
-│   ├── plugins-builtin/        # 随包发布的插件（diag-bridge · rag-bridge · serial-bridge）
+│   ├── mcp/                    # MCP 服务（25 个工具，进程内 workshop 面）
+│   ├── plugins-builtin/        # 随包发布的插件（serial-bridge · rag-bridge · diag-bridge · twin-injection-default）
 │   └── plugins/                # 运行时装配（单例）
+├── mcp/                        # aw mcp —— 零依赖 stdio MCP 网关（38 个工具，实例自动发现）
 ├── sdk/                        # agentworkshop/sdk——插件上下文、钩子总线、REST 客户端、浏览器 SDK
 ├── tui/                        # 终端工作台（aw tui）
 ├── shared/
-│   └── config/                 # schema.json（111 个设置描述符）+ 引擎（合并/校验/持久化）+ 路径解析器
+│   └── config/                 # schema.json（113 个设置描述符）+ 引擎（合并/校验/持久化）+ 路径解析器
 ├── config.yml                  # 工厂默认值（构建/启动时读取；版本号来自 package.json）
 ├── .AgentWorkShop/             # 检出内的配置根——prompts（版本化）+ 运行时覆盖 · 数据 · 日志 · 指令（git 忽略）
 ├── data/                       # 迁移前的旧版位置（自动迁移进配置根）
@@ -719,7 +729,7 @@ cd docs/site && npx vitepress build      # 生产构建 → .vitepress/dist
 | 插件系统 v2：浏览器面板注入、插件级 settings/groups、插件 i18n、宿主运行时服务（`ctx.services`/`ctx.daq`/`ctx.omp`） | 已交付 |
 | Instrument Glass 材质层（v0.7.36）：三级半透明材质、通透感、高光边缘、弹簧动效、逐页路由转场 | 已交付 |
 | 实时管线优化：数采帧索引 O(n²)→O(n)、逐 Agent 增量事件索引、图表原地更新、按大小感知的 JSON 持久化 | 已交付 |
-| AML 自动建模实验室：数据集构建 · 作业编排（uv 托管 Python）· 排行榜 · 晋级门禁 · 10 个 Agent 工具 | 已交付 |
+| AML 自动建模实验室：数据集构建 · 作业编排（uv 托管 Python）· 排行榜 · 晋级门禁 · 14 个 Agent 工具 | 已交付 |
 | Claude Agent SDK 适配器——常驻会话、同轮 steer 与 `canUseTool` HITL | 已交付 |
 | 工艺参数语义映射：`param_control`/`param_read` 按工艺参数寻址、四层写入限界、调控闭环（`dcw_judge`/`dcw_rollback`） | 已交付 |
 | Channel 群聊：请求升级为可追踪作业；原生 HITL（`omp` ask → 审批 → 回执） | 已交付 |
@@ -730,6 +740,12 @@ cd docs/site && npx vitepress build      # 生产构建 → .vitepress/dist
 | 运行时可观测：`GET /api/system/monitor` 团队指标，每项带文档化回滚开关 | 已交付 |
 | 工业基准：integrated 档（83 项检查、硬门禁）+ 多场景闭环（注塑 / 污水 / 退火 / BOPET）+ 质量目标 optloop 寻优 | 已交付 |
 | 生产构建全覆盖验收波次矩阵（约 1100+ 条真实断言，2026-09-24） | 已交付 |
+| 孪生 Provider 插件 SDK：物理/场景/目标/训练/求解器 Provider 经 `ctx.twin` 热加载注册，按 generation 装载 + 租约排空，内置 Provider 走同一外部契约 | 已交付 |
+| 任意场景混合孪生建模闭环：场景发现 → 编译 → 冻结 → 物理 spec 起草/校验/编译 → 参考训练器作业（物理校准 + 残差集成 + conformal UQ）、模型驱动试验、持续校准 worker | 已交付 |
+| 双模式优化闭环：`aml-training` / `aml-optimization` 专属频道模板、互斥工具面、探索 ↔ 生产双模式、带回读配对的治理微写 | 已交付 |
+| 产线操作流水：`/operations` 单页 + 工作台控制台面板，九类操作时间线、结构化写控 diff、闭环阶段推断 | 已交付 |
+| 外部 MCP 网关：`aw mcp` stdio 服务（38 个工具）、实例自动发现、`mcp.enabled` 运行时门控 | 已交付 |
+| 可配置系统时区（`time.timeZone`，live 热生效）+ 配方参数级单步限幅 `stepLimit` + 配方 `description` 元字段 | 已交付 |
 | 生产硬化：TLS、MQTT 鉴权、OPC UA 签名+加密缺省、结构化审计日志 | 规划中 |
 | 边缘部署形态：独立 edge-agent + 中心 broker | 规划中 |
 | 报警外送（邮件/webhook）+ 确认工作流 | 规划中 |

package/README.md

@@ -28,7 +28,7 @@
 
 **[简体中文](./README-zh.md)** · **[Documentation](https://kingdol666.github.io/AgentWorkShop)** · **[Releases](https://github.com/kingdol666/AgentWorkShop/releases)** · **[Changelog](./changelog.md)** · **[Plugin API](./docs/plugins.md)** · **[SDK](./docs/sdk.md)**
 
-<sub><b>v0.7.49</b> · 14 engines · 6 field protocols (5 built-in + serial plugin) · 111 runtime settings · bilingual docs (简体中文 / English)</sub>
+<sub><b>v0.7.50</b> · 14 engines · 6 field protocols (5 built-in + serial plugin) · 113 runtime settings · bilingual docs (简体中文 / English)</sub>
 
 <br />
 
@@ -79,7 +79,7 @@ Supervisory layer, second-level soft real-time<br/>
 7-state task machine · root queue + lease fencing<br/>
 Group chat jobs · scheduled tasks · HITL<br/>
 FTS5 + vector memory · hot-reloadable plugins<br/>
-SDK · CLI · TUI · ~1100+ acceptance assertions
+SDK · CLI · TUI · external MCP · ~1100+ acceptance assertions
 
 </td>
 </tr>
@@ -96,7 +96,7 @@ Docs are bilingual — the [VitePress site](https://kingdol666.github.io/AgentWo
 | **Getting started** | [Guide](https://kingdol666.github.io/AgentWorkShop/guide/getting-started) | `docs/site/guide/` | install → first run → first agent × line session, configuration, DAQ protocols, DCW read/write, HITL, recipe versioning, multi-harness teams, line permissions, AML |
 | **Plugin development** | [Plugin guide](https://kingdol666.github.io/AgentWorkShop/plugins/) | [`docs/plugins.md`](./docs/plugins.md) | the full extension contract: three scopes, `index.mjs` manifest, `ctx` server/browser surface, `settings`/`groups` declarations, lifecycle events, i18n, panels, team-scoped switches, a real example |
 | **SDK** | [SDK guide](https://kingdol666.github.io/AgentWorkShop/sdk/) | [`docs/sdk.md`](./docs/sdk.md) | `agentworkshop/sdk` as a programming client (platform REST client with envelope handling) and as the plugin extension base; TypeScript types; browser-side SDK |
-| **CLI** | [CLI manual](https://kingdol666.github.io/AgentWorkShop/cli/) | [`docs/cli.md`](./docs/cli.md) | all 14 `aw` commands, global options, exit codes, dual-mode path model, the command-registration system, maintainer publish guide |
+| **CLI** | [CLI manual](https://kingdol666.github.io/AgentWorkShop/cli/) | [`docs/cli.md`](./docs/cli.md) | all 15 `aw` commands, global options, exit codes, dual-mode path model, the command-registration system, maintainer publish guide |
 | **AML** | [AML guide](https://kingdol666.github.io/AgentWorkShop/guide/aml) | [`docs/aml.md`](./docs/aml.md) | auto-modeling lab: dataset build, job orchestration, model registry, promotion gates, agent tools |
 | **TUI** | — | [`docs/tui.md`](./docs/tui.md) · [`tui/README.md`](./tui/README.md) | terminal workbench: channels, agents, tasks, live monitor, HITL answering |
 | **Multi-harness architecture** | — | [`docs/multi-harness-architecture.md`](./docs/multi-harness-architecture.md) | engine taxonomy, normalized session contract, provider/model catalog, failure modes |
@@ -179,10 +179,14 @@ Recorded against a running instance: real DAQ history, real write control, real
 | **Full-operation audit log** | Every user / agent / system action lands in one queryable log; operators are attributed to "Channel/Member", distinct from users and the system. Streamed live over WS. |
 | **Team-scoped plugin switches** | Each team (channel) keeps an **independent plugin switch set** (`channel_plugins`): a disabled plugin's tools are not injected into that team's agents. Plugins themselves are hot-managed via `aw plugin` and the `/plugins` page. |
 | **Plugin extension API** | A self-contained directory under `plugins/<name>/` enhances **both halves at once**: `index.mjs` (server: hooks, routes, agent tools, **DAQ read drivers / DCW write drivers**, frame processors, node templates, config groups, KV, timers) and `client.mjs` (browser: panels injected into named slots, i18n, settings UI). Three scopes — `builtin` (shipped) > `project` (checkout) > `user` (`~/.AgentWorkShop`) — with ~1 s hot reload on enable/disable **and on code edits**; a disabled plugin's drivers are removed on hot reload immediately. Built-in example: **serial-bridge** (serial communication: read/write drivers, serial probe API, frontend panel). Full contract in [`docs/plugins.md`](./docs/plugins.md). |
-| **AML — auto-modeling lab** | Dataset build → training job → leaderboard → promotion gates → model reference, all driven from `/aml` or by agents through 10 `aml_*` tools. Python runtime bootstrapped with `uv` into an `./aml` asset root; artifacts and metadata stay under the config root. |
-| **AML hybrid twin × MPC** | Grey-box physics core + bounded PyTorch residual, `TwinSnapshot` / `VirtualTrial` / `RecommendationCertificate` with UQ/OOD screening and all-trajectory gates; the `hybrid_twin` channel profile injects 6 extra tools (`twin_*`, `mpc_optimize`). Trials stay virtual (`candidateExecuted=false`) and writes are recommendation-only until a certificate is certified. |
+| **AML — auto-modeling lab** | Dataset build → training job → leaderboard → promotion gates → model reference, all driven from `/aml` or by agents through **14 `aml_*` tools**. Python runtime bootstrapped with `uv` into an `aml/` asset root inside the config root; artifacts and metadata stay under the config root. Models carry a human-readable identity — label (line · recipe · objective), description and line/objective ids — selectable via `aml_model_find`. |
+| **AML hybrid twin × MPC** | Grey-box physics core + bounded PyTorch residual, `TwinSnapshot` / `VirtualTrial` / `RecommendationCertificate` with UQ/OOD screening and all-trajectory gates; the `hybrid_twin` channel profile injects **12 profile tools** (scene discovery/compile/freeze, physics-spec draft/validate/compile, snapshot, trial, MPC, gate, provider catalog) plus the general `twin_calibration_request`. Trials stay virtual (`candidateExecuted=false`) and writes are recommendation-only until a certificate is certified. |
+| **Two-mode optimization loop** | Dedicated channel templates split the loop **structurally**: an *aml-training* channel gets the modeling face (datasets, jobs, scene compile, physics specs, training plans — no plant writes, no MPC) and an *aml-optimization* channel gets the acting face (snapshots, trials, gate, MPC, Bayesian search `twin_bayes_optimize`, governed micro-writes `optimization_explore`) — training-while-optimizing is impossible by construction. **Exploration** mode writes through the governed chain with response read-back pairing (every write lands in an `optimization_explorations` ledger); **production** mode requires a bound model that cleared the twin gates (`aml` mode, `boundModelId` at instantiate time). |
 | **Runtime observability** | `GET /api/system/monitor` exposes the agent-team internals as numbers: root queue depth, watchdog interventions, harness session reuse, pending memory outbox — each guarded by a documented rollback switch, so new machinery can be turned off without a redeploy. |
-| **Fully config-driven runtime** | Every runtime knob (memory budgets, compaction, rollback guardrails, retention, backups, log level…) is declared once in the settings descriptor registry with precedence **config.yml < runtime-settings < env** — **111 settings across 16 groups** (32 live / 79 restart), no hardcoded defaults in code. |
+| **Line operations feed (`/operations`)** | One page for every touch on the plant: a live operation feed (REST snapshot + WS stream) classifies writes, DAQ, AML, recipe, rollback, alarm, line, manual and system actions into **nine categories**, renders structured **before → after diff badges** on write rows (failures marked red), and infers the **closed-loop phase** (monitor → exploration · small-step → recommendation search → shadow verification → production) straight from the event stream. The same feed powers a line-operations panel inside the workshop console. |
+| **External MCP gateway** | `aw mcp` starts a zero-dependency **stdio MCP server** (38 tools) that discovers a running instance automatically (env → lock file → config file) and exposes lines, recipes (incl. mark-good), governed DCW writes, semantic parameter mapping, DAQ queries, channels, team provisioning, model promotion, judge/rollback and plugin management to any MCP client — gated by the `mcp.enabled` runtime setting, with `--doctor` and `--print-config` helpers. |
+| **Fully config-driven runtime** | Every runtime knob (memory budgets, compaction, rollback guardrails, retention, backups, log level, system timezone, MCP gate…) is declared once in the settings descriptor registry with precedence **config.yml < runtime-settings < env** — **113 settings across 18 groups** (34 live / 79 restart), no hardcoded defaults in code. |
+| **Configurable system timezone** | One `time.timeZone` setting (IANA id, live-applied) drives human-facing timestamps — logs, chat, audit, agent records and schedules — through a single formatter; machine timelines (DAQ samples, `timestamptz`) stay absolute. |
 | **Configurable cadences** | Sampling and query defaults/floors are **live settings** (`daq.sampling.*`, `daq.query.*`): hot-reloaded, clamped on node create/patch, and agent tool descriptions always carry the current values. |
 
 #### Digital twin
@@ -250,7 +254,7 @@ page headers stack, dense tables become scrollable ledgers with a pinned identit
 <td width="50%"><img src="https://raw.githubusercontent.com/kingdol666/AgentWorkShop/main/docs/readme-assets/shot-monitor.png" alt="Runtime monitor" width="100%" /><br/><sub><b>Runtime monitor.</b> Every wired channel, member count, dependency cycle and owner.</sub></td>
 </tr>
 <tr>
-<td><img src="https://raw.githubusercontent.com/kingdol666/AgentWorkShop/main/docs/readme-assets/shot-settings.png" alt="Settings" width="100%" /><br/><sub><b>Settings.</b> 111 keys across 16 groups, descriptor-driven — the same registry the CLI reads.</sub></td>
+<td><img src="https://raw.githubusercontent.com/kingdol666/AgentWorkShop/main/docs/readme-assets/shot-settings.png" alt="Settings" width="100%" /><br/><sub><b>Settings.</b> 113 keys across 18 groups, descriptor-driven — the same registry the CLI reads.</sub></td>
 <td><img src="https://raw.githubusercontent.com/kingdol666/AgentWorkShop/main/docs/readme-assets/shot-plugins.png" alt="Plugins" width="100%" /><br/><sub><b>Plugins.</b> Three scopes, hot reload on edit, per-team switches.</sub></td>
 </tr>
 </table>
@@ -373,7 +377,7 @@ aw update --check                      # only report; nothing is installed
 npm install -g agentworkshop@latest    # manual equivalent
 ```
 
-Releases follow semver. `aw start` verifies the config root on every launch and migrates the legacy pre-`home` `data/` layout into it (newest file wins), so data survives upgrades. SQLite schema migrations run server-side at boot. Current version: **v0.7.49** — see [Releases](https://github.com/kingdol666/AgentWorkShop/releases).
+Releases follow semver. `aw start` verifies the config root on every launch and migrates the legacy pre-`home` `data/` layout into it (newest file wins), so data survives upgrades. SQLite schema migrations run server-side at boot. Current version: **v0.7.50** — see [Releases](https://github.com/kingdol666/AgentWorkShop/releases).
 
 ### Your first agent × line session (~2 minutes)
 
@@ -422,6 +426,7 @@ Agent tool descriptions are re-rendered with the current values on every injecti
 | `aw stop` | Stop a running `aw` service instance via the single-instance lock |
 | `aw config list · get · set · unset · reset` | Read & write runtime settings (validated against the schema, atomic writes) |
 | `aw plugin list · create · enable · disable` | Manage plugins across all three scopes: inspect, scaffold (project scope; `--global` for user scope), enable/disable (state file, ~1 s hot reload on the running server) |
+| `aw mcp` | Industrial MCP stdio server (38 tools over the platform REST: lines, recipes, DCW writes, parameter mapping, DAQ, channels, teams, model promotion, plugins) — auto-discovers the running instance; `--doctor` diagnoses discovery/health/auth, `--print-config` emits a ready-to-paste MCP client config |
 | `aw home` | Inspect / initialize the config root `.AgentWorkShop` |
 | `aw init <dir>` | Scaffold a runnable project (full config system + CLI included) |
 | `aw register <path\|url\|npm:pkg>` | Register a new command — project-local or `--global` |
@@ -482,10 +487,14 @@ export async function run(argv, ctx) {
 
 The modeling half of the loop: `/aml` builds **datasets** out of tagged telemetry, submits **training jobs** to a `uv`-managed Python runtime, ranks runs on a **leaderboard**, and promotes a model only when it clears the configured **gates** (NRMSE, rollout NRMSE, validation/test gap, minimum rows and runs). Promoted models are referenced by id, so an MPC or shadow-twin controller can consume a versioned artifact instead of a vague "latest".
 
-- Agents do the same work through 10 tools (`aml_dataset_build`, `aml_job_submit`, `aml_job_status`, `aml_leaderboard`, `aml_model_promote`, …) — submit a goal and let the team train and report.
+- Agents do the same work through 14 tools (`aml_dataset_build`, `aml_job_submit`, `aml_job_status`, `aml_leaderboard`, `aml_model_promote`, `aml_model_find`, `aml_training_plan_*`, …) — submit a goal and let the team train and report.
 - The Python runtime bootstraps itself with `uv` into an `./aml` asset root inside the config root; dataset/artifact/metadata paths, disk quota, job timeout, concurrency and retention are all **settings** (16 in the `aml` group), not constants.
 - Governance defaults are conservative: cross-recipe datasets are refused unless explicitly allowed, and every job is attributed.
-- **Hybrid twin × MPC core**: AML also owns the *hybrid twin* plane — `SceneContract` / `PhysicsModelManifest` / `TwinSnapshot` / `ObjectiveProfile` / `VirtualTrial` / `RecommendationCertificate` on top of a grey-box physics core with a bounded PyTorch residual, UQ/OOD screening and all-trajectory hard-constraint gates. Channels opt in through the `hybrid_twin` profile, which injects six extra agent tools (`twin_scene_read`, `twin_snapshot_create`, `twin_trial_run`, `mpc_optimize`, `twin_gate_evaluate`, `twin_calibration_request`); a trial is *virtual* by design (`candidateExecuted=false`) and a recommendation is only certified when gates and benefit pass. Server-side strategies are `safe_small_step` (insufficient data) and `precise_search` (model cleared gates). Plan + acceptance record: [`docs/aml-hybrid-twin-mpc-integration-plan.md`](./docs/aml-hybrid-twin-mpc-integration-plan.md), [`docs/aml-hybrid-twin-implementation-acceptance.md`](./docs/aml-hybrid-twin-implementation-acceptance.md).
+- **Hybrid twin × MPC core**: AML also owns the *hybrid twin* plane — `SceneContract` / `PhysicsModelManifest` / `TwinSnapshot` / `ObjectiveProfile` / `VirtualTrial` / `RecommendationCertificate` on top of a grey-box physics core with a bounded PyTorch residual, UQ/OOD screening and all-trajectory hard-constraint gates. Channels opt in through the `hybrid_twin` profile, which injects twelve profile agent tools (`twin_provider_catalog`, `twin_scene_discover/compile/freeze`, `twin_physics_spec_draft/validate/compile`, `twin_scene_read`, `twin_snapshot_create`, `twin_trial_run`, `mpc_optimize`, `twin_gate_evaluate`) plus the general `twin_calibration_request`; a trial is *virtual* by design (`candidateExecuted=false`) and a recommendation is only certified when gates and benefit pass. Server-side strategies are `safe_small_step` (insufficient data) and `precise_search` (model cleared gates). Plan + acceptance record: [`docs/aml-hybrid-twin-mpc-integration-plan.md`](./docs/aml-hybrid-twin-mpc-integration-plan.md), [`docs/aml-hybrid-twin-implementation-acceptance.md`](./docs/aml-hybrid-twin-implementation-acceptance.md).
+- **Any-scene modeling loop**: the loop is not tied to a bundled scene. Bind nodes → `twin_scene_discover` infers control/state/disturbance/target/guard semantics with evidence → `twin_scene_compile` drafts a `SceneContract` → you confirm → `twin_scene_freeze` freezes it (immutable versions) → `twin_physics_spec_draft` proposes a skeleton PhysicsSpec (one first-order relaxation equation per state, τ/gain as calibratable prior-boxed parameters) that you refine → validate/compile (whitelisted AST, no arbitrary code) → `aml_job_submit { job_kind: 'hybrid_residual' }` runs without any code: the built-in reference trainer calibrates physics parameters (stage A), fits a 3-member bounded residual ensemble (stage B) and reports conformal q90 coverage (stage D) — the calibrated spec is written back and shared by evaluation, prediction and trials.
+- **Model-backed twin & continuous calibration**: `twin_trial_run` / `mpc_optimize` accept a `model_id` — the rollout is then driven by the trained hybrid model (calibrated physics + residual-ensemble ONNX, `ModelBackedHybridProvider`) with UQ coverage and member disagreement coming from training artifacts, and a dataset whose line/product/recipe doesn't match the scene is rejected (`TWIN_MODEL_SCENE_MISMATCH`). Calibration requests are consumed by a 60 s worker that re-submits `hybrid_residual` retraining *candidates* automatically; promotion still goes through HITL + Twin Gate (`AML_TWIN_WORKER_DISABLED=1` turns the worker off).
+- **Two-mode optimization loop**: `aw` ships two channel templates — `chtpl-aml-training-default` (modeling face: 22 tools, no plant writes, no MPC) and `chtpl-aml-optimization-default` (acting face: snapshot/trial/gate/MPC/Bayesian search plus `optimization_explore`, the governed micro-write that pairs every setpoint with its response read-back in an `optimization_explorations` ledger). The faces are mutually exclusive per channel; optimization channels instantiate either in **exploration** mode (governed small steps, no model required) or **aml** mode (`boundModelId` required — a gate-cleared model must be bound before the acting face unlocks).
+- **Twin provider plugin SDK (hot-loaded)**: an industrial scene's physics core / scene packs / objectives / solvers register through `ctx.twin` into the **Twin Provider Registry** (`registerPhysicsProvider` / `registerScenePack` / `registerObjectiveProfile` / `registerTrainingAdapter` / `registerSolverAdapter`, plus `listProviders` / `getProviderHealth` / `resolveProvider` / `validateProvider` / `retireProvider` / `isRegistryAvailable`). Twin tools resolve providers via `SceneContract.physicsProfileId` (falling back to the bundled `twin-injection-default`), so **swapping scenes changes neither tools nor channels**; a hot reload stages a new **generation** and drains the old one while trials and MPC hold leases, so in-flight work is never invalidated and `retireProvider` only unloads at zero references. The bundled injection provider ships through that same external contract (`server/plugins-builtin/twin-injection-default/`), and plugins live in the usual scopes with no restart for disable/reload. Field tables and a runnable example: [`docs/plugins.md`](./docs/plugins.md) → "Hybrid-twin provider plugins"; SDK surface: [`docs/sdk.md`](./docs/sdk.md) §4.10.
 
 ---
 
@@ -600,12 +609,12 @@ in [`docs/audit/e2e-2026-09-24-all-features.md`](./docs/audit/e2e-2026-09-24-all
 | Real-engine collaboration | task-flow real **PASS**, `e2e-omp-workspace` **13 / 13**, group-chat real **PASS**, multi-user HIL **PASS** (2026-09-24) | real `omp` lead + workers: decomposition, dispatch, progress, artifacts, workspace-scoped file reads, unload; multi-user HITL visibility; multi-harness team 14/18 (omp + dsh green) | `npx tsx scripts/e2e-agentteam-task-flow-real.ts` · `npx tsx scripts/e2e-omp-workspace.ts` |
 | Front-end acceptance | responsive gate **16 / 16** routes, nav drawer **10 / 10** (2026-09-24, real Chromium) | every product page incl. the 3D town renders at the gate thresholds; drawer interaction, scroll lock, route change, zero page errors | `node scripts/ui/verify-responsive.mjs --routes all` · `node scripts/ui/verify-nav.mjs` |
 | `_aw0746-stage12.mjs` | **10 / 0** (2026-09-25, globally installed **v0.7.49 packaged build**) | health gate (version=0.7.49) → admin login/role surface → channel creation with a real `omp` lead → group chat opened → **real omp reply** → omp child process visible in monitoring | `AW_BASE=<base> node scripts/_aw0746-stage12.mjs` |
-| `_aw0746-stage34.mjs` | **11 / 0** (2026-09-25, v0.7.49 packaged build + PLC simulator) | line provisioned from the simulator (11 DCW + 14 DAQ) → product/recipe triple → **every sample tagged with recipe_id/run_id** while running → `daq_query` **defaults to the active batch** (`scope=all` opts out, measured 1 vs 4) → Hybrid Twin template instantiated (lead + 3 workers, 6 twin tools injected, 57 total) | `AW_BASE=<base> node scripts/_aw0746-stage34.mjs` |
+| `_aw0746-stage34.mjs` | **13 / 0** (2026-09-25, v0.7.49 packaged build + PLC simulator) | line provisioned from the simulator (11 DCW + 14 DAQ) → product/recipe triple → **every sample tagged with recipe_id/run_id** while running → `daq_query` **defaults to the active batch** (`scope=all` opts out, measured 1 vs 4) → Hybrid Twin template instantiated (lead + 3 workers, 6 twin tools injected, 57 total) | `AW_BASE=<base> node scripts/_aw0746-stage34.mjs` |
 | `_aw0746-stage6b-train-sweep.mjs` | gates **all green** (2026-09-25, v0.7.46 packaged build, real torch training) | 1 s grid dataset (**1724 rows / 10 runs**) → inline-code job trained for real → G1 one-step **0.0992** (≤0.10) / G2 rollout 0.1092 (≤0.25) / G3 generalisation 14.6% (≤20%) / G4 rows+runs → model registered and **promoted to production** | `AW_BASE=<base> AW_DATASET_ID=<ds> node scripts/_aw0746-stage6b-train-sweep.mjs` |
 | `_aw0746-stage7-twin.mjs` | **17 / 0** (2026-09-25, v0.7.49 packaged build, real DAQ + real model) | real samples → TwinSnapshot `fresh=true` → 65 real VirtualTrials (12 safe candidates, `candidateExecuted=false`) → 12-check recommendation gate **write_eligible** → verdict written back to the model as `twinEligibility` → `mpc_optimize` upgrades to **precise_search** → **0 DCW writes** throughout | `AW_BASE=<base> AW_REQUIRE_MODEL=1 node scripts/_aw0746-stage7-twin.mjs` |
 | `_aw0746-stage6c-jobsmoke.mjs` | **5 / 0** (2026-09-25, v0.7.49 packaged build) | AML job path: inline `code` accepted (no `workspace/train.py` required) → job reaches a terminal state → training logs/ONNX traces → experiment plus per-check gates persisted | `AW_BASE=<base> node scripts/_aw0746-stage6c-jobsmoke.mjs` |
 | `_aw0748-failclosed-probe.mjs` | **7 / 0** (2026-09-25, v0.7.49 packaged build) | hard-constraint **fail-closed** regression: `auto_daq` builds a snapshot from real bindings → a mappable `weight` constraint evaluates normally → an unmappable `part_weight` constraint is judged not passed with a precise detail (v0.7.47 passed it falsely) → an empty trajectory is also judged not passed | `AW_BASE=<base> node scripts/_aw0748-failclosed-probe.mjs` |
-| `_aw0748-mock-closedloop.mjs` | **11 / 0** (2026-09-25, v0.7.49 packaged build, in-process `mock`, no quota) | scheduler closed-loop regression: create a mock channel → the root task is genuinely dispatched and closed → `[mock:complex]` forces delegation into child tasks → the parent closes after they finish → **the second and third root tasks are dispatched too** (before the fix the lead was unloaded while idle, `cr.scheduler` became null and later root tasks only landed in the DB, sitting in SUBMITTED until `ROOT_TIMEOUT`) → the tool bridge explicitly rejects host tools for mock | `AW_BASE=<base> node scripts/_aw0748-mock-closedloop.mjs` |
+| `_aw0748-mock-closedloop.mjs` | **13 / 0** (2026-09-25, v0.7.49 packaged build, in-process `mock`, no quota) | scheduler closed-loop regression: create a mock channel → the root task is genuinely dispatched and closed → `[mock:complex]` forces delegation into child tasks → the parent closes after they finish → **the second and third root tasks are dispatched too** (before the fix the lead was unloaded while idle, `cr.scheduler` became null and later root tasks only landed in the DB, sitting in SUBMITTED until `ROOT_TIMEOUT`) → the tool bridge explicitly rejects host tools for mock | `AW_BASE=<base> node scripts/_aw0748-mock-closedloop.mjs` |
 | Industrial bench — integrated | **83 / 83 checks, hard gate green** (2026-09-21) | five-protocol five-line bench: connectivity, sampling into Timescale, governed writes, governance, agent closed loop, semantic parameter layer | `node bench/pipeline.mjs --profile integrated` |
 | Multi-scenario closed loop | 4 scenarios · closed loops reached (2026-09-21) | injection-molding weight-window tuning · A²/O wastewater compliance with energy minimum · continuous-annealing quality window vs throughput · BOPET line mission | `node bench/scenarios.mjs --scenarios injection,wwtp,anneal` |
 | Line permissions / audit-negative | 21 / 21 + 9 / 9 (2026-09-12) | three-state line grants with human-readable 403s, binding-subject validation, revocation convergence, unauthenticated WS receives zero telemetry | `node scripts/_dbg-perms-e2e.mjs <base> <adminPass>` · `node scripts/_dbg-audit-neg-e2e.mjs <base> <adminPass>` |
@@ -628,7 +637,7 @@ AgentWorkShop/
 │   ├── pages/                  # / · /workshop · /workshop/agents · /workshop/teams
 │   │                           # /workshop/channel-templates · /workshop/schedules · /workshop/w/:id
 │   │                           # /town · /daq · /daq/:id · /dcw · /dcw/:id
-│   │                           # /aml · /monitor · /logs · /permissions
+│   │                           # /aml · /operations · /monitor · /logs · /permissions
 │   │                           # /plugins · /users · /tokens · /settings
 │   ├── components/workshop/    # timeline · lanes · task board · memory panel · 3D town
 │   └── stores/composables/     # Pinia + AEP client
@@ -639,13 +648,14 @@ AgentWorkShop/
 │   │   ├── agents/             # AgentInterface: 14 engines (+ industrial tools)
 │   │   ├── daq/ dcw/ aml/      # edge runtimes · drivers · bus · storage · modeling lab
 │   │   └── db/                 # repos over node:sqlite
-│   ├── mcp/                    # MCP server (25 tools)
-│   ├── plugins-builtin/        # plugins shipped with the package (diag-bridge · rag-bridge · serial-bridge)
+│   ├── mcp/                    # MCP server (25 tools, in-process workshop face)
+│   ├── plugins-builtin/        # plugins shipped with the package (serial-bridge · rag-bridge · diag-bridge · twin-injection-default)
 │   └── plugins/                # runtime assembly (singletons)
+├── mcp/                        # aw mcp — zero-dependency stdio MCP gateway (38 tools, instance auto-discovery)
 ├── sdk/                        # agentworkshop/sdk — plugin context, hook bus, REST client, browser SDK
 ├── tui/                        # terminal workbench (aw tui)
 ├── shared/
-│   └── config/                 # schema.json (111 setting descriptors) + engine (merge/validate/persist) + path resolver
+│   └── config/                 # schema.json (113 setting descriptors) + engine (merge/validate/persist) + path resolver
 ├── config.yml                  # factory defaults (read at build/start; version comes from package.json)
 ├── .AgentWorkShop/             # config root in a checkout — prompts (versioned) + runtime overrides · data · logs · commands (git-ignored)
 ├── data/                       # legacy pre-migration location (auto-migrated into the config root)
@@ -740,6 +750,12 @@ building, so those files are the single source of truth for the single-page guid
 | Runtime observability: `GET /api/system/monitor` team metrics, each with a documented rollback switch | Shipped |
 | Industrial bench: integrated profile (83 checks, hard gate) + multi-scenario closed loops (injection / wwtp / anneal / BOPET) + quality-target optloop | Shipped |
 | Full-coverage acceptance wave matrix on a production build (~1100+ real assertions, 2026-09-24) | Shipped |
+| Twin provider plugin SDK: hot-loadable physics/scene/objective/training/solver providers via `ctx.twin`, generation-staged reload with lease draining, bundled provider shipped through the same external contract | Shipped |
+| Any-scene hybrid-twin modeling loop: scene discover → compile → freeze → physics-spec draft/validate/compile → reference-trainer job (physics calibration + residual ensemble + conformal UQ), model-backed trials, continuous-calibration worker | Shipped |
+| Two-mode optimization loop: dedicated `aml-training` / `aml-optimization` channel templates, mutually exclusive tool faces, exploration ↔ production modes with governed read-back-paired micro-writes | Shipped |
+| Line operations feed: `/operations` page + workshop-console panel, nine-category operation timeline, structured write diffs, closed-loop phase inference | Shipped |
+| External MCP gateway: `aw mcp` stdio server (38 tools), instance auto-discovery, `mcp.enabled` runtime gate | Shipped |
+| Configurable system timezone (`time.timeZone`, live) + recipe `stepLimit` per-parameter single-step bound + recipe `description` metadata | Shipped |
 | Production hardening: TLS, MQTT auth, OPC UA sign+encrypt defaults, structured audit log | Planned |
 | Edge deployment shape: standalone edge-agent + central broker | Planned |
 | Alarm outbound delivery (email/webhook) + ack workflow | Planned |

package/app/components/AppSidebar.vue

@@ -52,6 +52,7 @@ const menuGroups = computed<MenuGroup[]>(() => [
     id: 'ops',
     label: t('menu.groups.ops'),
     items: [
+      { key: '/operations', icon: 'i-tabler-stream', label: t('menu.operations'), motion: 'im-pop' },
       { key: '/monitor', icon: 'i-tabler-cpu', label: t('menu.monitor'), motion: 'im-pulse' },
       { key: '/logs', icon: 'i-tabler-list-details', label: t('menu.logs'), motion: 'im-pop' },
     ],