@loom-framework/core 0.1.0-alpha.177 → 0.1.0-alpha.179

package/builtin-skills/app-skill/references/model.md

@@ -0,0 +1,124 @@
+# Model CLI Reference
+
+## Overview
+
+Model management commands configure AI models used by the Loom platform — for chat, Process execution, and AI-assisted features.
+
+## Commands
+
+### `loom model list`
+
+List all configured AI models with their capability tier, tags, and default status.
+
+```bash
+loom model list
+```
+
+Output shows each model's name, capability tier (`[fast]`, `[standard]`, `[max]`), default marker, base URL, auth status, and tags.
+
+### `loom model add <name>`
+
+Add a new AI model configuration.
+
+```bash
+loom model add "claude-sonnet-4-6" \
+  --base-url "https://api.anthropic.com" \
+  --auth-token "sk-ant-..." \
+  --default \
+  --tier standard \
+  --tags "创作模型,多模态"
+```
+
+**Options:**
+| Flag | Required | Description |
+|------|----------|-------------|
+| `--base-url` | Yes | API base URL |
+| `--auth-token` | No | Authentication token |
+| `--default` | No | Set as default model (first model is auto-default) |
+| `--tier` | No | Capability tier: `fast`, `standard`, or `max` |
+| `--tags` | No | Comma-separated classification tags |
+| `--token` | No | Auth token (or set LOOM_TOKEN) |
+| `--server` | No | Server URL (or set LOOM_SERVER, default: http://localhost:3000) |
+
+### `loom model update <name>`
+
+Update an existing model configuration.
+
+```bash
+# Change capability tier
+loom model update "claude-sonnet-4-6" --tier fast
+
+# Set as default
+loom model update "claude-opus-4-7" --default
+
+# Update tags
+loom model update "claude-sonnet-4-6" --tags "创作模型,多模态,代码"
+
+# Clear capability tier
+loom model update "claude-sonnet-4-6" --tier none
+
+# Clear tags
+loom model update "claude-sonnet-4-6" --tags ""
+```
+
+**Options:**
+| Flag | Required | Description |
+|------|----------|-------------|
+| `--base-url` | No | Update API base URL |
+| `--auth-token` | No | Update authentication token |
+| `--default` | No | Set as default model |
+| `--tier` | No | Update capability tier (`fast`/`standard`/`max`, or `none` to clear) |
+| `--tags` | No | Update tags (comma-separated, or `""` to clear) |
+| `--token` | No | Auth token (or set LOOM_TOKEN) |
+| `--server` | No | Server URL (or set LOOM_SERVER, default: http://localhost:3000) |
+
+### `loom model remove <name>`
+
+Remove a model configuration. If the removed model was the default, the first remaining model becomes the new default.
+
+```bash
+loom model remove "old-model"
+```
+
+## Capability Tiers
+
+Capability tiers are informational labels for model capability:
+
+| Tier | Meaning | Use Case |
+|------|---------|----------|
+| `fast` | Lightweight, quick responses | Simple data aggregation, formatting |
+| `standard` | Balanced performance | General-purpose tasks |
+| `max` | Strongest reasoning | Complex analysis, multi-step reasoning |
+
+Capability tiers appear as colored tags in the AI Sidecar and ChatDrawer model selectors, and in Process overview when a model is bound.
+
+## Process Model Binding
+
+When creating a Process, you can specify which model it uses:
+
+```bash
+# Use a specific model for complex reasoning
+loom process add --name "daily-analysis" \
+  --cron "0 2 * * *" \
+  --prompt "Analyze daily data and generate insights" \
+  --owner admin \
+  --model "claude-opus-4-7"
+
+# Use a fast model for simple aggregation
+loom process add --name "quick-summary" \
+  --cron "*/30 * * * *" \
+  --prompt "Summarize key metrics" \
+  --owner admin \
+  --model "claude-sonnet-4-6"
+```
+
+If `--model` is omitted, the Process uses the global default model.
+
+## HTTP API Equivalents
+
+| CLI Command | HTTP Endpoint |
+|------------|---------------|
+| `loom model list` | `GET /api/v1/ai/models` |
+| `loom model add` | `POST /api/v1/ai/models` |
+| `loom model update` | `PUT /api/v1/ai/models/:name` |
+| `loom model remove` | `DELETE /api/v1/ai/models/:name` |

package/builtin-skills/app-skill/references/process-builder.md

@@ -26,6 +26,7 @@ Identify:
 - **What** the process should do
 - **When** it should run (schedule or event trigger)
 - **Who** owns it (username for service token)
+- **Which model** should run it (use `loom model list` to see available models; fast for simple tasks, max for complex reasoning; omit to use global default)
 - **Complexity** (lightweight vs full)
 
 ### 2. Choose Trigger
@@ -45,7 +46,8 @@ For lightweight:
 loom process add --name "task-summary" \
   --cron "0 9 * * *" \
   --prompt "Read all tasks created yesterday, summarize key themes, and post a digest" \
-  --owner admin
+  --owner admin \
+  --model "claude-sonnet-4-6"
 ```
 
 For full:
@@ -170,6 +172,7 @@ _Only create the `references/` directory and add files when needed._
 - Use `loom data` commands for data operations, not direct file I/O
 - Include error handling in Decision Points
 - Add a notification step at the end of Flow to send execution results to yourself: `loom notification send --userId <owner> --type success --title "Process completed" --description "Summary of results" --token $LOOM_TOKEN`
+- If your process needs strong reasoning, specify a max-level model via `--model` when creating it. Fast models are best for simple data aggregations.
 
 ## Metrics Definition
 

package/builtin-skills/app-skill/references/process.md

@@ -37,6 +37,13 @@ loom process add --name "health-check" \
   --prompt "Check system health" \
   --metrics-config ./metrics-config.json \
   --owner bot
+
+# With a specific model
+loom process add --name "complex-analysis" \
+  --cron "0 2 * * *" \
+  --prompt "Analyze daily data and generate insights" \
+  --owner admin \
+  --model "claude-opus-4-7"
 ```
 
 **Options:**
@@ -55,6 +62,7 @@ loom process add --name "health-check" \
 | `--owner-token` | No | Direct JWT token (bypasses runtime resolution) |
 | `--metrics-config` | No | Path to metrics DashboardWidget config JSON |
 | `--source` | No | Feeder script command (auto-registers lifecycle script + event subscription) |
+| `--model` | No | AI model name for this process (must exist in model management; use `loom model list` to see available models) |
 | `--token` | No | Auth token (or set LOOM_TOKEN) |
 | `--server` | No | Server URL (or set LOOM_SERVER, default: http://localhost:3000) |
 
@@ -121,6 +129,7 @@ loom process update daily-summary --description "Updated description" --cron "0
 | `--event-debounce` | No | Update debounce window in ms |
 | `--prompt` | No | Update inline prompt |
 | `--description` | No | Update description |
+| `--model` | No | Update AI model name (pass empty string to remove model binding) |
 | `--token` | No | Auth token (or set LOOM_TOKEN) |
 | `--server` | No | Server URL (or set LOOM_SERVER, default: http://localhost:3000) |
 

package/builtin-skills/loom/SKILL.md

@@ -17,7 +17,7 @@ Loom 是**配置驱动的 AI 平台开发框架**——你描述需求，AI 通
 
 典型使用方式：
 
-1. **用自然语言描述需求** — 告诉 Claude Code（或其它支持Skill的 Agent） 你要做什么（如"用 loom 开发一个错题管理平台"），Claude Code 会调用本 Skill 读取配置、生成页面、启动服务
+1. **用自然语言描述需求** — 告诉 Claude Code（或其它支持Skill的 Agent） 你要做什么（如"用 loom 开发一个告警管理平台"），Claude Code 会交互式地向你确认完整需求，然后调用本 Skill 读取配置、生成页面、启动服务
 2. **对话迭代** — 通过Claude Code 或 创建的项目 AI 对话面板改需求、加字段、调样式，AI 理解 loom 的工作方式，知道何时改配置、何时跑命令、何时直接改代码
 3. **日常运营** — 通过 AI 对话面板或 `loom data` CLI 查看、创建、更新数据
 
@@ -96,7 +96,7 @@ loom generate capabilities
 
 每次修改数据模型或 AI 按钮后，重新运行此命令即可更新 `references/` 下的自动生成内容，`SKILL.md` 中手动补充的内容不会被覆盖。
 
-首次生成后，需要补充 `SKILL.md` 主文档中的 TODO 内容，AI 对话时才能正确理解项目业务语义：
+首次生成项目SKILL后，需要补充其 `SKILL.md` 主文档中的 TODO 内容，AI 对话时才能正确理解项目业务语义：
 
 1. **frontmatter `description`**：写入用户在 AI 对话中会说的触发短语。格式参考：
    ```
@@ -161,7 +161,7 @@ loom dev --frontend-port 8080     # 临时覆盖前端端口
 
 **`--force` 重建（默认）**：字段变更时直接 `--force` 重建页面，是最快最干净的方式，页面与配置立即同步。`--force` 会自动将旧页面备份到 `.loom/backup/<时间戳>/` 目录，如需恢复直接复制回来即可。
 
-**AI 增量更新（有定制时）**：如果用户说"保留我的定制"或明确表示页面有手动修改过，才让 AI 增量更新。告诉 AI 你改了什么（如"我在配置里加了一个学期字段"），AI 会精确添加新字段的列和表单项，保留定制代码。如果AI进行了定制化开发，务必将定制化开发内容更新到CLAUDE.md，使得后续新启动会话时Claude Code可掌握项目定制化情况。
+**AI 增量更新（有定制时）**：如果用户说"保留我的定制"或明确表示页面有手动修改过，才让 AI 增量更新。告诉 AI 你改了什么（如"我在配置里加了一个严重等级字段"），AI 会精确添加新字段的列和表单项，保留定制代码。如果AI进行了定制化开发，务必将定制化开发内容更新到CLAUDE.md，使得后续新启动会话时Claude Code可掌握项目定制化情况。
 
 ```bash
 # 只改了 AI 按钮或 enum → 重启后端即可
@@ -255,8 +255,8 @@ export default defineConfig({
   {
     id: 'analyze',            // 带变量的示例
     label: { 'zh-CN': '分析', 'en-US': 'Analyze' },
-    prompt: '分析{{questionContent}}，错误答案：{{wrongAnswer}}',
-    placement: 'wrong_questions',  // 可选：限制按钮只出现在指定模型页面，逗号分隔多个模型名；省略则出现在所有页面
+    prompt: '分析告警原因：{{description}}，当前状态：{{status}}',
+    placement: 'alerts',  // 可选：限制按钮只出现在指定模型页面，逗号分隔多个模型名；省略则出现在所有页面
   }],
   dashboards: [{              // 可选：数据概览，widget/图表类型/筛选配置详见 references/dashboard.md
     name: 'overview',         // 必填：唯一标识
@@ -330,8 +330,8 @@ export default defineConfig({
 
 所有面向用户的展示字段均支持 `LocaleString` 格式：`string | Record<string, string>`。
 
-- **简单写法**：`description: '科目'` — 所有语言显示相同文本
-- **多语言写法**：`description: { 'zh-CN': '科目', 'en-US': 'Subject' }` — 切换语言时自动翻译
+- **简单写法**：`description: '级别'` — 所有语言显示相同文本
+- **多语言写法**：`description: { 'zh-CN': '级别', 'en-US': 'Level' }` — 切换语言时自动翻译
 - **enum 值**：`enum` 始终是存储值（不分语言），显示文本通过 `enumLabels` 翻译
 - **回退规则**：请求的 locale → `'zh-CN'` → 第一个可用值
 

package/builtin-skills/loom/references/README.md

@@ -22,15 +22,15 @@ Loom 项目的主要交互方式是**和 AI 对话**。你告诉 AI 想要什么
 
 ## 第一个项目
 
-假设你要做一个**错题管理平台**。
+假设你要做一个**告警管理平台**。
 
 ### 第 1 步：告诉 AI 你的想法
 
 在 Claude Code 中说：
 
-> 用 loom 创建一个错题管理平台，项目名 wrong-questions，filesystem 适配器。
-> 需要两个模型：错题记录（科目、题目内容、错误答案、正确答案、错误类型、是否掌握）和复习计划（错题ID、计划日期、状态）。
-> AI 按钮要能分析错因和出类似题。
+> 用 loom 创建一个告警管理平台，项目名 alert-manager，filesystem 适配器。
+> 需要两个模型：告警记录（级别、描述、来源、严重度、状态）和处理记录（告警ID、处理人、处理时间、状态）。
+> AI 按钮要能分析告警根因和生成处置建议。
 
 AI 会读取 Loom Skill，自动执行：
 
@@ -51,9 +51,9 @@ loom dev
 ```
 
 打开 http://localhost:5173 ，你的平台已经能用了：
-- 顶部导航有「错题记录」和「复习计划」页面
+- 顶部导航有「告警记录」和「处理记录」页面
 - 每个页面有新增、编辑、删除、筛选
-- 错题记录每行有 AI 按钮：分析错因、出类似题
+- 告警记录每行有 AI 按钮：分析根因、生成处置建议
 - 右侧 AI 面板可以自然语言操作数据
 
 ### 第 3 步：继续和 AI 聊
@@ -62,11 +62,11 @@ loom dev
 
 | 你说的 | AI 做的 |
 |-------|--------|
-| "帮我记一道数学错题，3+5=7，正确答案是8" | 用 `loom data write` 创建记录 |
-| "看看哪些错题还没掌握" | 查数据并展示 |
+| "帮我记一条告警，服务器CPU过载，当前状态未处理" | 用 `loom data write` 创建记录 |
+| "看看哪些告警还没处理" | 查数据并展示 |
 | "把这个页面改成卡片布局" | 修改页面 TSX |
-| "加一个学期字段" | 改 loom.config.ts + 改页面 + 更新 Skill references |
-| "加一个科目分布的饼图" | 修改 Dashboard 页面 |
+| "加一个严重等级字段" | 改 loom.config.ts + 改页面 + 更新 Skill references |
+| "加一个级别分布的饼图" | 修改 Dashboard 页面 |
 
 这些操作在终端 Claude Code 和页面对话框里都能做。
 
@@ -81,7 +81,7 @@ loom dev
 | 新增数据模型 | 需要生成新页面 | 改配置 → `loom generate capabilities` + `loom generate page <Name> --model <model>` → 重启后端 |
 | 新增/删除字段 | 需要 | 改配置 → AI 更新页面（推荐）或 `--force` 重建 → `loom generate capabilities` 更新 Skill |
 
-**推荐：AI 辅助迭代**。告诉 AI 你改了什么（比如"我在配置里加了一个学期字段"），AI 会在页面中精确添加新字段的列和表单项，**保留你之前做的视觉定制代码**。
+**推荐：AI 辅助迭代**。告诉 AI 你改了什么（比如"我在配置里加了一个严重等级字段"），AI 会在页面中精确添加新字段的列和表单项，**保留你之前做的视觉定制代码**。
 
 **谨慎使用 `--force`**。它会用最新配置全量覆盖页面，之前对页面做的任何定制修改都会丢失。
 
@@ -89,7 +89,7 @@ loom dev
 # 只改了 AI 按钮或 enum → 重启后端即可，不用碰页面
 # 加了新字段 → 让 AI 更新页面
 # 想重来一个干净页面 → --force
-loom generate page WrongQuestions --model wrong_questions --force
+loom generate page Alerts --model alerts --force
 ```
 
 页面不存在时直接 `loom generate page` 即可，不需要 `--force`。
@@ -150,22 +150,22 @@ export default defineConfig({
 
 每个项目的 `.claude/skills/<项目名>/SKILL.md` 是 AI 理解你业务的入口。`loom generate capabilities` 生成骨架（`references/` 下的内容自动维护），但 SKILL.md 主文档需要你补充三个地方：
 
-1. **description**：写用户会说的话，如 `"记录错题", "查看错题", "分析错因"`
+1. **description**：写用户会说的话，如 `"记录告警", "查看告警", "分析根因"`
 2. **Overview**：2-3 句话描述平台做什么
 3. **Usage Scenarios**：5-10 个典型用户请求及对应的 `loom data` 命令
 
-补充后，AI 在终端和对话框里都能理解"帮我查数学错题"这类业务请求。
+补充后，AI 在终端和对话框里都能理解"帮我查未处理告警"这类业务请求。
 
 ## 数据操作
 
 不打开页面也能操作数据。你可以自己跑命令，也可以让 AI 跑：
 
 ```bash
-loom data read wrong_questions --filter '{"subject":"数学"}' --limit 10
-loom data write wrong_questions --data '{"subject":"数学","questionContent":"3+5=?","wrongAnswer":"7","correctAnswer":"8"}'
-loom data update wrong_questions --id rec_xxx --data '{"isMastered":true}'
-loom data delete wrong_questions --id rec_xxx
-loom data schema wrong_questions    # 查看字段结构
+loom data read alerts --filter '{"level":"critical"}' --limit 10
+loom data write alerts --data '{"level":"critical","description":"CPU使用率超过95%","source":"monitor","status":"open"}'
+loom data update alerts --id rec_xxx --data '{"status":"resolved"}'
+loom data delete alerts --id rec_xxx
+loom data schema alerts    # 查看字段结构
 ```
 
 ## 后端扩展

package/builtin-skills/loom/references/dashboard.md

@@ -47,7 +47,7 @@ export default defineConfig({
       field: 'score',        // aggregate 为 sum/avg/min/max 时必填：数值字段名
       filter: { status: 'active' },  // 可选：聚合前筛选条件
       groupBy: 'category',   // 可选：分组字段（string/enum/date/boolean）
-      crossGroupBy: 'grade', // 可选：第二分组字段，用于二维图表
+      crossGroupBy: 'source', // 可选：第二分组字段，用于二维图表
       interval: 'month',      // 可选：date 字段分组粒度 minute|hour|day|week|month（默认 month）
       dateField: 'createdAt', // 可选：单组件日期筛选字段，覆盖 dashboard 级 defaultDateField
     }]
@@ -61,7 +61,7 @@ export default defineConfig({
 |------|------|------|
 | `aggregate` | `count\|sum\|avg\|min\|max\|ratio` | 聚合方式。`ratio` = 符合 filter 条件的记录数/总记录数，用于 gauge/liquid/stat 百分比 |
 | `field` | string | 聚合目标字段。仅 aggregate 为 sum/avg/min/max 时需要 |
-| `filter` | object | 筛选条件，键值对。如 `{isMastered: true}` 筛选已掌握的记录 |
+| `filter` | object | 筛选条件，键值对。如 `{isResolved: true}` 筛选已处理的记录 |
 | `groupBy` | string | 主分组字段。string/enum 字段按值分组计数；boolean 字段自动映射为"是/否"；date 字段按 interval 分段 |
 | `crossGroupBy` | string | 第二分组维度。配合 groupBy 使用，产生交叉分组数据，用于二维图表 |
 | `interval` | `minute\|hour\|day\|week\|month` | date 字段的分组粒度，仅 groupBy 指向 date 字段时有效 |
@@ -81,7 +81,7 @@ export default defineConfig({
 ```typescript
 {
   field: 'subject',           // 必填：筛选字段名
-  label: { 'zh-CN': '科目' }, // 必填：显示标签，支持多语言
+  label: { 'zh-CN': '级别' }, // 必填：显示标签，支持多语言
   type: 'select',             // 必填：当前仅支持 'select'（'dateRange' 已定义但暂未实现）
   options: ['语文', '数学'],   // 可选：select 类型的选项。省略则从数据中自动提取去重值
 }
@@ -94,11 +94,11 @@ export default defineConfig({
 ```typescript
 dashboards: [{
   name: 'overview',
-  models: ['wrong_questions'],
+  models: ['alerts'],
   defaultDateField: 'createdAt',          // 启用日期范围筛选
   filterDimensions: [                      // 启用 select 下拉筛选（日期范围用 defaultDateField）
-    { field: 'subject', label: { 'zh-CN': '科目', 'en-US': 'Subject' }, type: 'select' },
-    { field: 'difficulty', label: { 'zh-CN': '难度', 'en-US': 'Difficulty' }, type: 'select' },
+    { field: 'subject', label: { 'zh-CN': '级别', 'en-US': 'Level' }, type: 'select' },
+    { field: 'difficulty', label: { 'zh-CN': '严重度', 'en-US': 'Severity' }, type: 'select' },
   ],
   layout: [/* ... */],
 }]
@@ -111,28 +111,28 @@ dashboards: [{
 | type | 中文名称 | 必填字段 | 典型配置 | 生成效果 |
 |------|---------|---------|---------|---------|
 | `stat` | 统计卡片 | — | `{ aggregate: "count" }` 或 `{ aggregate: "avg", field: "score" }` 或 `{ aggregate: "ratio", filter: { passed: true } }` | Ant Design Statistic 数值卡片 |
-| `gauge` | 仪表盘 | `aggregate: "ratio"` + `filter` | `{ aggregate: "ratio", filter: { isMastered: true } }` | G2 仪表盘（百分比进度） |
-| `liquid` | 水波图 | `aggregate: "ratio"` + `filter` | `{ aggregate: "ratio", filter: { isMastered: true } }` | G2 水波图（百分比进度） |
+| `gauge` | 仪表盘 | `aggregate: "ratio"` + `filter` | `{ aggregate: "ratio", filter: { isResolved: true } }` | G2 仪表盘（百分比进度） |
+| `liquid` | 水波图 | `aggregate: "ratio"` + `filter` | `{ aggregate: "ratio", filter: { isResolved: true } }` | G2 水波图（百分比进度） |
 
-> **说明**：gauge 和 liquid 使用 `aggregate: "ratio"` + `filter` 来计算百分比（如掌握率 = 已掌握数/总数），ratio 自动算出 0-1 之间的值。
+> **说明**：gauge 和 liquid 使用 `aggregate: "ratio"` + `filter` 来计算百分比（如处理率 = 已处理数/总数），ratio 自动算出 0-1 之间的值。
 
 #### 占比类 — 展示分布构成
 
 | type | 中文名称 | 必填字段 | 典型配置 | 生成效果 |
 |------|---------|---------|---------|---------|
 | `pie` | 饼图 | `groupBy` (enum/string) | `{ groupBy: "subject" }` | G2 饼图（分布占比） |
-| `ring` | 环形图 | `groupBy` (boolean/enum) | `{ groupBy: "difficulty" }` 或 `{ groupBy: "isMastered" }` | G2 环形图（占比对比，boolean 自动映射"是/否"） |
+| `ring` | 环形图 | `groupBy` (boolean/enum) | `{ groupBy: "difficulty" }` 或 `{ groupBy: "isResolved" }` | G2 环形图（占比对比，boolean 自动映射"是/否"） |
 | `treemap` | 矩形树图 | `groupBy` + 可选 `crossGroupBy` | `{ groupBy: "subject" }` 或 `{ groupBy: "subject", crossGroupBy: "difficulty" }` | G2 矩形树图（层级占比，有 crossGroupBy 则两层） |
 
 #### 对比类 — 跨类别比较
 
 | type | 中文名称 | 必填字段 | 典型配置 | 生成效果 |
 |------|---------|---------|---------|---------|
-| `bar` | 条形图 | `groupBy` | `{ groupBy: "errorType" }` | G2 条形图（横向分类对比） |
-| `stacked_bar` | 堆叠条形图 | `groupBy` + `crossGroupBy` | `{ groupBy: "subject", crossGroupBy: "errorType" }` | G2 堆叠条形图（组成+对比） |
+| `bar` | 条形图 | `groupBy` | `{ groupBy: "source" }` | G2 条形图（横向分类对比） |
+| `stacked_bar` | 堆叠条形图 | `groupBy` + `crossGroupBy` | `{ groupBy: "subject", crossGroupBy: "source" }` | G2 堆叠条形图（组成+对比） |
 | `grouped_bar` | 分组条形图 | `groupBy` + `crossGroupBy` | `{ groupBy: "subject", crossGroupBy: "difficulty" }` | G2 分组条形图（并列对比） |
 | `radar` | 雷达图 | `groupBy` | `{ groupBy: "subject" }` | G2 雷达图（多维度闭合多边形） |
-| `funnel` | 漏斗图 | `groupBy` | `{ groupBy: "errorType" }` | G2 漏斗图（阶段转化，柱条宽度反映数值） |
+| `funnel` | 漏斗图 | `groupBy` | `{ groupBy: "source" }` | G2 漏斗图（阶段转化，柱条宽度反映数值） |
 
 #### 趋势类 — 时间维度变化
 
@@ -146,7 +146,7 @@ dashboards: [{
 | type | 中文名称 | 必填字段 | 典型配置 | 生成效果 |
 |------|---------|---------|---------|---------|
 | `scatter` | 散点图 | `groupBy` | `{ groupBy: "subject" }` | G2 散点图（相关性/分布） |
-| `heatmap` | 热力图 | `groupBy` + `crossGroupBy` | `{ groupBy: "subject", crossGroupBy: "grade" }` | G2 热力图（密度/交叉分析，颜色深浅表示数量） |
+| `heatmap` | 热力图 | `groupBy` + `crossGroupBy` | `{ groupBy: "subject", crossGroupBy: "source" }` | G2 热力图（密度/交叉分析，颜色深浅表示数量） |
 
 ## 执行 generate dashboard
 
@@ -171,63 +171,63 @@ cd frontend && pnpm add @antv/g2
 
 ## 完整配置示例
 
-以小学生错题管理平台为例，展示所有 15 种图表的配置：
+以小学生告警管理平台为例，展示所有 15 种图表的配置：
 
 ```typescript
 import { defineConfig } from '@loom-framework/core';
 
 export default defineConfig({
-  project: { name: 'wrong-questions' },
+  project: { name: 'alert-manager' },
   data: { /* ... */ },
   dashboards: [{
     name: 'overview',
-    description: { 'zh-CN': '错题数据概览', 'en-US': 'Overview' },
-    models: ['wrong_questions', 'review_plans'],
+    description: { 'zh-CN': '告警数据概览', 'en-US': 'Overview' },
+    models: ['alerts', 'alert_handlers'],
     defaultDateField: 'createdAt',
     filterDimensions: [
-      { field: 'subject', label: { 'zh-CN': '科目', 'en-US': 'Subject' }, type: 'select' },
-      { field: 'difficulty', label: { 'zh-CN': '难度', 'en-US': 'Difficulty' }, type: 'select' },
+      { field: 'subject', label: { 'zh-CN': '级别', 'en-US': 'Level' }, type: 'select' },
+      { field: 'difficulty', label: { 'zh-CN': '严重度', 'en-US': 'Severity' }, type: 'select' },
     ],
     layout: [
       {
         row: [
-          { type: 'stat', title: { 'zh-CN': '错题总数', 'en-US': 'Total Questions' }, model: 'wrong_questions', aggregate: 'count', span: 4 },
-          { type: 'stat', title: { 'zh-CN': '掌握率', 'en-US': 'Mastery Rate' }, model: 'wrong_questions', aggregate: 'ratio', filter: { isMastered: true }, span: 4 },
-          { type: 'gauge', title: { 'zh-CN': '掌握率仪表盘', 'en-US': 'Mastery Gauge' }, model: 'wrong_questions', aggregate: 'ratio', filter: { isMastered: true }, span: 6 },
-          { type: 'liquid', title: { 'zh-CN': '掌握率水波图', 'en-US': 'Mastery Liquid' }, model: 'wrong_questions', aggregate: 'ratio', filter: { isMastered: true }, span: 5 },
-          { type: 'stat', title: { 'zh-CN': '平均复习次数', 'en-US': 'Avg Reviews' }, model: 'wrong_questions', aggregate: 'avg', field: 'reviewCount', span: 5 }
+          { type: 'stat', title: { 'zh-CN': '告警总数', 'en-US': 'Total Alerts' }, model: 'alerts', aggregate: 'count', span: 4 },
+          { type: 'stat', title: { 'zh-CN': '处理率', 'en-US': 'Resolution Rate' }, model: 'alerts', aggregate: 'ratio', filter: { isResolved: true }, span: 4 },
+          { type: 'gauge', title: { 'zh-CN': '处理率仪表盘', 'en-US': 'Resolution Gauge' }, model: 'alerts', aggregate: 'ratio', filter: { isResolved: true }, span: 6 },
+          { type: 'liquid', title: { 'zh-CN': '处理率水波图', 'en-US': 'Resolution Liquid' }, model: 'alerts', aggregate: 'ratio', filter: { isResolved: true }, span: 5 },
+          { type: 'stat', title: { 'zh-CN': '平均处理时长', 'en-US': 'Avg Resolution Time' }, model: 'alerts', aggregate: 'avg', field: 'resolutionTime', span: 5 }
         ]
       },
       {
         row: [
-          { type: 'pie', title: { 'zh-CN': '科目分布', 'en-US': 'By Subject' }, model: 'wrong_questions', groupBy: 'subject', span: 8 },
-          { type: 'ring', title: { 'zh-CN': '难度分布', 'en-US': 'By Difficulty' }, model: 'wrong_questions', groupBy: 'difficulty', span: 8 },
-          { type: 'bar', title: { 'zh-CN': '错误类型分布', 'en-US': 'By Error Type' }, model: 'wrong_questions', groupBy: 'errorType', span: 8 }
+          { type: 'pie', title: { 'zh-CN': '级别分布', 'en-US': 'By Level' }, model: 'alerts', groupBy: 'subject', span: 8 },
+          { type: 'ring', title: { 'zh-CN': '严重度分布', 'en-US': 'By Severity' }, model: 'alerts', groupBy: 'difficulty', span: 8 },
+          { type: 'bar', title: { 'zh-CN': '告警来源分布', 'en-US': 'By Source' }, model: 'alerts', groupBy: 'source', span: 8 }
         ]
       },
       {
         row: [
-          { type: 'line', title: { 'zh-CN': '错题录入趋势', 'en-US': 'Questions Trend' }, model: 'wrong_questions', groupBy: 'createdAt', interval: 'month', span: 12 },
-          { type: 'area', title: { 'zh-CN': '复习计划趋势', 'en-US': 'Review Plans Trend' }, model: 'review_plans', groupBy: 'createdAt', interval: 'month', span: 12 }
+          { type: 'line', title: { 'zh-CN': '告警趋势', 'en-US': 'Alerts Trend' }, model: 'alerts', groupBy: 'createdAt', interval: 'month', span: 12 },
+          { type: 'area', title: { 'zh-CN': '处理记录趋势', 'en-US': 'Handlers Trend' }, model: 'alert_handlers', groupBy: 'createdAt', interval: 'month', span: 12 }
         ]
       },
       {
         row: [
-          { type: 'radar', title: { 'zh-CN': '各科目概览', 'en-US': 'Subject Overview' }, model: 'wrong_questions', groupBy: 'subject', span: 8 },
-          { type: 'stacked_bar', title: { 'zh-CN': '科目×错误类型', 'en-US': 'Subject × Error Type' }, model: 'wrong_questions', groupBy: 'subject', crossGroupBy: 'errorType', span: 8 },
-          { type: 'grouped_bar', title: { 'zh-CN': '科目×难度', 'en-US': 'Subject × Difficulty' }, model: 'wrong_questions', groupBy: 'subject', crossGroupBy: 'difficulty', span: 8 }
+          { type: 'radar', title: { 'zh-CN': '各级别概览', 'en-US': 'Level Overview' }, model: 'alerts', groupBy: 'subject', span: 8 },
+          { type: 'stacked_bar', title: { 'zh-CN': '级别×错误类型', 'en-US': 'Level × Error Type' }, model: 'alerts', groupBy: 'subject', crossGroupBy: 'source', span: 8 },
+          { type: 'grouped_bar', title: { 'zh-CN': '级别×严重度', 'en-US': 'Level × Severity' }, model: 'alerts', groupBy: 'subject', crossGroupBy: 'difficulty', span: 8 }
         ]
       },
       {
         row: [
-          { type: 'scatter', title: { 'zh-CN': '科目分布散点', 'en-US': 'Subject Scatter' }, model: 'wrong_questions', groupBy: 'subject', span: 8 },
-          { type: 'funnel', title: { 'zh-CN': '错误类型漏斗', 'en-US': 'Error Type Funnel' }, model: 'wrong_questions', groupBy: 'errorType', span: 8 },
-          { type: 'treemap', title: { 'zh-CN': '科目×难度树图', 'en-US': 'Subject × Difficulty Tree' }, model: 'wrong_questions', groupBy: 'subject', crossGroupBy: 'difficulty', span: 8 }
+          { type: 'scatter', title: { 'zh-CN': '级别分布散点', 'en-US': 'Level Scatter' }, model: 'alerts', groupBy: 'subject', span: 8 },
+          { type: 'funnel', title: { 'zh-CN': '告警来源漏斗', 'en-US': 'Source Funnel' }, model: 'alerts', groupBy: 'source', span: 8 },
+          { type: 'treemap', title: { 'zh-CN': '级别×严重度树图', 'en-US': 'Level × Severity Tree' }, model: 'alerts', groupBy: 'subject', crossGroupBy: 'difficulty', span: 8 }
         ]
       },
       {
         row: [
-          { type: 'heatmap', title: { 'zh-CN': '科目×年级热力图', 'en-US': 'Subject × Grade Heatmap' }, model: 'wrong_questions', groupBy: 'subject', crossGroupBy: 'grade', span: 24 }
+          { type: 'heatmap', title: { 'zh-CN': '级别×年级热力图', 'en-US': 'Level × Grade Heatmap' }, model: 'alerts', groupBy: 'subject', crossGroupBy: 'source', span: 24 }
         ]
       }
     ]
@@ -243,11 +243,11 @@ export default defineConfig({
 2. 使用 `/antv-g2-chart` skill 生成 G2 v5 Spec 代码
 3. 添加 `<DashboardChart spec={...} />` 组件
 
-示例——追加一个科目×年级的堆叠柱状图：
+示例——追加一个级别×年级的堆叠柱状图：
 
 ```tsx
 // 在已有的 Dashboard 页面中追加
-const subjectGradeData = useCrossGroupChartData(wrongQuestionsList, { groupBy: 'subject', crossGroupBy: 'grade' });
+const levelSourceData = useCrossGroupChartData(alertsList, { groupBy: 'level', crossGroupBy: 'source' });
 
 <DashboardChart spec={{
   type: 'interval',

package/builtin-skills/loom/references/data-model.md

@@ -79,7 +79,7 @@ aiButtons: [{
   label: { 'zh-CN': '总结', 'en-US': 'Summarize' },  // required: LocaleString, button text
   icon: 'FileTextOutlined', // optional: Ant Design icon component name
   prompt: '请总结以下内容',   // required: supports {{var}} placeholders, replaced with current row field values
-  placement: 'wrong_questions',  // optional: limit to specific model pages, comma-separated; omit for all pages
+  placement: 'alerts',  // optional: limit to specific model pages, comma-separated; omit for all pages
 }]
 ```
 
@@ -99,6 +99,10 @@ ai: {
         authToken: 'sk-xxx',       // optional: ANTHROPIC_AUTH_TOKEN, passed via --settings
         baseUrl: 'https://api.example.com',  // optional: ANTHROPIC_BASE_URL, passed via --settings
         isDefault: true,           // optional: default model for new chats
+        tier: "max",
+          tags: [
+            "自然语言"
+          ]
       },
     ],
   },
@@ -148,9 +152,9 @@ ui: {
 ```typescript
 relations: [{
   name: 'question_reviews',            // required: unique relation identifier
-  label: { 'zh-CN': '复习记录', 'en-US': 'Review Records' },  // optional: LocaleString
-  from: 'wrong_questions',             // required: source model name
-  to: 'review_plans',                  // required: target model name
+  label: { 'zh-CN': '处理记录', 'en-US': 'Handler Records' },  // optional: LocaleString
+  from: 'alerts',             // required: source model name
+  to: 'alert_handlers',                  // required: target model name
   foreignKey: 'questionId',            // required: foreign key field in target model
   type: 'hasMany',                     // required: 'hasOne' | 'hasMany' | 'manyToMany'
   through: 'enrollments',              // optional: pivot model for manyToMany
@@ -215,8 +219,8 @@ enumLabel(schema, 'status', 'active')  // → '激活'
 
 All user-facing string fields support `LocaleString` format: `string | Record<string, string>`.
 
-- **Simple**: `description: '科目'` — same text for all locales
-- **Multi-locale**: `description: { 'zh-CN': '科目', 'en-US': 'Subject' }` — auto-translated on locale switch
+- **Simple**: `description: '级别'` — same text for all locales
+- **Multi-locale**: `description: { 'zh-CN': '级别', 'en-US': 'Level' }` — auto-translated on locale switch
 - **Fallback**: requested locale → `'zh-CN'` → first available value
 
 Supported fields: `project.description`, `ModelSchema.description`, `FieldDefinition.description`, `AIButtonConfig.label`, `DashboardConfig.description`, `DashboardWidget.title`, `RelationConfig.label`, `DashboardFilterDimension.label`

package/dist/backend/process/engine.d.ts

@@ -49,7 +49,7 @@ export declare class ProcessEngine {
     removeProcess(name: string): Promise<boolean>;
     pauseProcess(name: string): Promise<void>;
     resumeProcess(name: string): Promise<void>;
-    updateProcess(name: string, updates: Partial<Pick<ProcessDefinition, 'description' | 'cron' | 'cronTimezone' | 'onEvent' | 'eventFilter' | 'eventDebounceMs' | 'prompt' | 'paused' | 'metricsConfig' | 'source'>>): Promise<ProcessDefinition | undefined>;
+    updateProcess(name: string, updates: Partial<Pick<ProcessDefinition, 'description' | 'cron' | 'cronTimezone' | 'onEvent' | 'eventFilter' | 'eventDebounceMs' | 'prompt' | 'paused' | 'metricsConfig' | 'source' | 'model'>>): Promise<ProcessDefinition | undefined>;
     runProcess(name: string, triggerType: 'cron' | 'event' | 'manual', eventPayload?: Record<string, unknown>): Promise<string>;
     /** Build AI prompt from process definition */
     buildProcessPrompt(definition: ProcessDefinition, runId: string, triggerType: 'cron' | 'event' | 'manual', eventPayload?: Record<string, unknown>): Promise<{

package/dist/backend/process/engine.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"engine.d.ts","sourceRoot":"","sources":["../../../src/backend/process/engine.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAIH,OAAO,KAAK,EAAE,QAAQ,EAAmC,MAAM,mBAAmB,CAAC;AAEnF,OAAO,KAAK,EAAE,UAAU,EAAE,MAAM,qBAAqB,CAAC;AACtD,OAAO,KAAK,EACV,eAAe,EACf,iBAAiB,EACjB,aAAa,EACb,iBAAiB,EACjB,eAAe,EACf,uBAAuB,EACxB,MAAM,wBAAwB,CAAC;AAMhC,OAAO,EAAE,aAAa,EAAE,MAAM,aAAa,CAAC;AAE5C,OAAO,KAAK,EAAE,qBAAqB,EAAE,wBAAwB,EAAE,MAAM,wBAAwB,CAAC;AAC9F,OAAO,EAAE,SAAS,EAAE,MAAM,uBAAuB,CAAC;AAElD,OAAO,KAAK,EAAE,QAAQ,EAAE,MAAM,wBAAwB,CAAC;AACvD,OAAO,KAAK,EAAE,aAAa,EAAE,MAAM,6BAA6B,CAAC;AACjE,OAAO,KAAK,EAAE,cAAc,EAAgC,MAAM,0BAA0B,CAAC;AAE7F,OAAO,KAAK,EAAE,SAAS,EAAE,MAAM,kBAAkB,CAAC;AAIlD,MAAM,WAAW,oBAAoB;IACnC,MAAM,EAAE,eAAe,CAAC;IACxB,MAAM,EAAE,QAAQ,CAAC;IACjB,cAAc,EAAE,cAAc,CAAC;IAC/B,KAAK,EAAE,SAAS,CAAC;IACjB,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB,QAAQ,CAAC,EAAE,QAAQ,CAAC;IACpB,aAAa,CAAC,EAAE,aAAa,CAAC;IAC9B,UAAU,CAAC,EAAE,UAAU,CAAC;IACxB,SAAS,CAAC,EAAE,SAAS,CAAC;IACtB,SAAS,CAAC,EAAE,MAAM,CAAC;CACpB;AAED,qBAAa,aAAa;IACxB,OAAO,CAAC,QAAQ,CAAkB;IAClC,OAAO,CAAC,cAAc,CAAiB;IACvC,OAAO,CAAC,KAAK,CAAiB;IAC9B,OAAO,CAAC,MAAM,CAAgB;IAC9B,OAAO,CAAC,YAAY,CAAsB;IAC1C,OAAO,CAAC,MAAM,CAAW;IACzB,OAAO,CAAC,cAAc,CAAiB;IACvC,OAAO,CAAC,MAAM,CAAkB;IAChC,OAAO,CAAC,KAAK,CAAY;IACzB,OAAO,CAAC,aAAa,CAAS;IAC9B,OAAO,CAAC,QAAQ,CAAC,CAAW;IAC5B,OAAO,CAAC,aAAa,CAAC,CAAgB;IACtC,OAAO,CAAC,UAAU,CAAC,CAAa;IAChC,OAAO,CAAC,SAAS,CAAC,CAAY;IAC9B,OAAO,CAAC,SAAS,CAAC,CAAS;IAC3B,OAAO,CAAC,UAAU,CAAS;gBAEf,OAAO,EAAE,oBAAoB;IAuBnC,UAAU,IAAI,OAAO,CAAC,IAAI,CAAC;IA4BjC,OAAO,CAAC,qBAAqB;IAOvB,UAAU,CAAC,GAAG,EAAE,IAAI,CAAC,iBAAiB,EAAE,WAAW,GAAG,WAAW,GAAG,QAAQ,CAAC,GAAG,OAAO,CAAC,iBAAiB,CAAC;IAmC1G,aAAa,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,OAAO,CAAC;IAc7C,YAAY,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC;IAUzC,aAAa,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC;IAc1C,aAAa,CAAC,IAAI,EAAE,MAAM,EAAE,OAAO,EAAE,OAAO,CAAC,IAAI,CAAC,iBAAiB,EAAE,aAAa,GAAG,MAAM,GAAG,cAAc,GAAG,SAAS,GAAG,aAAa,GAAG,iBAAiB,GAAG,QAAQ,GAAG,QAAQ,GAAG,eAAe,GAAG,QAAQ,CAAC,CAAC,GAAG,OAAO,CAAC,iBAAiB,GAAG,SAAS,CAAC;IAyB1P,UAAU,CAAC,IAAI,EAAE,MAAM,EAAE,WAAW,EAAE,MAAM,GAAG,OAAO,GAAG,QAAQ,EAAE,YAAY,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,GAAG,OAAO,CAAC,MAAM,CAAC;IAuBjI,8CAA8C;IACxC,kBAAkB,CACtB,UAAU,EAAE,iBAAiB,EAC7B,KAAK,EAAE,MAAM,EACb,WAAW,EAAE,MAAM,GAAG,OAAO,GAAG,QAAQ,EACxC,YAAY,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,GACrC,OAAO,CAAC;QAAE,MAAM,EAAE,MAAM,CAAC;QAAC,KAAK,CAAC,EAAE,KAAK,CAAC;YAAE,IAAI,EAAE,MAAM,CAAC;YAAC,IAAI,EAAE,MAAM,CAAA;SAAE,CAAC,CAAA;KAAE,CAAC;IAiE7E,+EAA+E;IAC/E,OAAO,CAAC,yBAAyB;IAcjC,iFAAiF;IACjF,OAAO,CAAC,oBAAoB;IA8B5B;;;;OAIG;YACW,eAAe;IAmD7B,OAAO,CAAC,cAAc;YAYR,YAAY;IAyK1B,cAAc,IAAI;QAAE,MAAM,EAAE,MAAM,CAAC;QAAC,OAAO,EAAE,MAAM,CAAC;QAAC,aAAa,EAAE,MAAM,CAAC;QAAC,OAAO,EAAE,iBAAiB,EAAE,CAAA;KAAE;IAI1G,gBAAgB,CAAC,IAAI,EAAE,MAAM,GAAG,aAAa;IAe7C,aAAa,CAAC,OAAO,CAAC,EAAE;QAAE,MAAM,CAAC,EAAE,QAAQ,GAAG,QAAQ,CAAA;KAAE,GAAG,OAAO,CAAC,iBAAiB,EAAE,CAAC;IAIvF,UAAU,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,iBAAiB,GAAG,SAAS,CAAC;IAIhE,UAAU,IAAI,iBAAiB,EAAE;IAIjC,SAAS,IAAI,aAAa;IAIpB,OAAO,CAAC,WAAW,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE;QAAE,KAAK,CAAC,EAAE,MAAM,CAAC;QAAC,MAAM,CAAC,EAAE,MAAM,CAAC;QAAC,IAAI,CAAC,EAAE,MAAM,CAAC;QAAC,EAAE,CAAC,EAAE,MAAM,CAAA;KAAE,GAAG,OAAO,CAAC,eAAe,EAAE,CAAC;IAInI,WAAW,CAAC,WAAW,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE;QAAE,IAAI,CAAC,EAAE,MAAM,CAAC;QAAC,EAAE,CAAC,EAAE,MAAM,CAAA;KAAE,GAAG,OAAO,CAAC,MAAM,CAAC;IAI3F,MAAM,CAAC,WAAW,EAAE,MAAM,EAAE,KAAK,EAAE,MAAM,GAAG,OAAO,CAAC,eAAe,GAAG,IAAI,CAAC;IAI3E,UAAU,CAAC,WAAW,EAAE,MAAM,GAAG,OAAO,CAAC,uBAAuB,EAAE,CAAC;IAMnE,YAAY,CAAC,WAAW,EAAE,MAAM,EAAE,QAAQ,EAAE,IAAI,CAAC,qBAAqB,EAAE,WAAW,CAAC,GAAG;QAAE,SAAS,CAAC,EAAE,MAAM,CAAA;KAAE,GAAG,OAAO,CAAC,IAAI,CAAC;IAQ7H,kBAAkB,CAAC,WAAW,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE;QAAE,UAAU,CAAC,EAAE,MAAM,CAAC;QAAC,IAAI,CAAC,EAAE,MAAM,CAAC;QAAC,EAAE,CAAC,EAAE,MAAM,CAAC;QAAC,KAAK,CAAC,EAAE,MAAM,CAAA;KAAE,GAAG,OAAO,CAAC,qBAAqB,EAAE,CAAC;IAIxJ,oBAAoB,CAAC,WAAW,EAAE,MAAM,GAAG,OAAO,CAAC,wBAAwB,GAAG,IAAI,CAAC;IAInF,oBAAoB,CAAC,WAAW,EAAE,MAAM,EAAE,MAAM,EAAE,wBAAwB,GAAG,OAAO,CAAC,IAAI,CAAC;IAI1F,uBAAuB,CAAC,WAAW,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC;IAI3D,QAAQ,IAAI,OAAO,CAAC,IAAI,CAAC;CA8BhC"}
+{"version":3,"file":"engine.d.ts","sourceRoot":"","sources":["../../../src/backend/process/engine.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAIH,OAAO,KAAK,EAAE,QAAQ,EAAmC,MAAM,mBAAmB,CAAC;AAEnF,OAAO,KAAK,EAAE,UAAU,EAAE,MAAM,qBAAqB,CAAC;AACtD,OAAO,KAAK,EACV,eAAe,EACf,iBAAiB,EACjB,aAAa,EACb,iBAAiB,EACjB,eAAe,EACf,uBAAuB,EACxB,MAAM,wBAAwB,CAAC;AAMhC,OAAO,EAAE,aAAa,EAAE,MAAM,aAAa,CAAC;AAE5C,OAAO,KAAK,EAAE,qBAAqB,EAAE,wBAAwB,EAAE,MAAM,wBAAwB,CAAC;AAC9F,OAAO,EAAE,SAAS,EAAE,MAAM,uBAAuB,CAAC;AAElD,OAAO,KAAK,EAAE,QAAQ,EAAE,MAAM,wBAAwB,CAAC;AACvD,OAAO,KAAK,EAAE,aAAa,EAAE,MAAM,6BAA6B,CAAC;AACjE,OAAO,KAAK,EAAE,cAAc,EAAgC,MAAM,0BAA0B,CAAC;AAE7F,OAAO,KAAK,EAAE,SAAS,EAAE,MAAM,kBAAkB,CAAC;AAIlD,MAAM,WAAW,oBAAoB;IACnC,MAAM,EAAE,eAAe,CAAC;IACxB,MAAM,EAAE,QAAQ,CAAC;IACjB,cAAc,EAAE,cAAc,CAAC;IAC/B,KAAK,EAAE,SAAS,CAAC;IACjB,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB,QAAQ,CAAC,EAAE,QAAQ,CAAC;IACpB,aAAa,CAAC,EAAE,aAAa,CAAC;IAC9B,UAAU,CAAC,EAAE,UAAU,CAAC;IACxB,SAAS,CAAC,EAAE,SAAS,CAAC;IACtB,SAAS,CAAC,EAAE,MAAM,CAAC;CACpB;AAED,qBAAa,aAAa;IACxB,OAAO,CAAC,QAAQ,CAAkB;IAClC,OAAO,CAAC,cAAc,CAAiB;IACvC,OAAO,CAAC,KAAK,CAAiB;IAC9B,OAAO,CAAC,MAAM,CAAgB;IAC9B,OAAO,CAAC,YAAY,CAAsB;IAC1C,OAAO,CAAC,MAAM,CAAW;IACzB,OAAO,CAAC,cAAc,CAAiB;IACvC,OAAO,CAAC,MAAM,CAAkB;IAChC,OAAO,CAAC,KAAK,CAAY;IACzB,OAAO,CAAC,aAAa,CAAS;IAC9B,OAAO,CAAC,QAAQ,CAAC,CAAW;IAC5B,OAAO,CAAC,aAAa,CAAC,CAAgB;IACtC,OAAO,CAAC,UAAU,CAAC,CAAa;IAChC,OAAO,CAAC,SAAS,CAAC,CAAY;IAC9B,OAAO,CAAC,SAAS,CAAC,CAAS;IAC3B,OAAO,CAAC,UAAU,CAAS;gBAEf,OAAO,EAAE,oBAAoB;IAuBnC,UAAU,IAAI,OAAO,CAAC,IAAI,CAAC;IA4BjC,OAAO,CAAC,qBAAqB;IAOvB,UAAU,CAAC,GAAG,EAAE,IAAI,CAAC,iBAAiB,EAAE,WAAW,GAAG,WAAW,GAAG,QAAQ,CAAC,GAAG,OAAO,CAAC,iBAAiB,CAAC;IAmC1G,aAAa,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,OAAO,CAAC;IAc7C,YAAY,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC;IAUzC,aAAa,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC;IAc1C,aAAa,CAAC,IAAI,EAAE,MAAM,EAAE,OAAO,EAAE,OAAO,CAAC,IAAI,CAAC,iBAAiB,EAAE,aAAa,GAAG,MAAM,GAAG,cAAc,GAAG,SAAS,GAAG,aAAa,GAAG,iBAAiB,GAAG,QAAQ,GAAG,QAAQ,GAAG,eAAe,GAAG,QAAQ,GAAG,OAAO,CAAC,CAAC,GAAG,OAAO,CAAC,iBAAiB,GAAG,SAAS,CAAC;IAyBpQ,UAAU,CAAC,IAAI,EAAE,MAAM,EAAE,WAAW,EAAE,MAAM,GAAG,OAAO,GAAG,QAAQ,EAAE,YAAY,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,GAAG,OAAO,CAAC,MAAM,CAAC;IAuBjI,8CAA8C;IACxC,kBAAkB,CACtB,UAAU,EAAE,iBAAiB,EAC7B,KAAK,EAAE,MAAM,EACb,WAAW,EAAE,MAAM,GAAG,OAAO,GAAG,QAAQ,EACxC,YAAY,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,GACrC,OAAO,CAAC;QAAE,MAAM,EAAE,MAAM,CAAC;QAAC,KAAK,CAAC,EAAE,KAAK,CAAC;YAAE,IAAI,EAAE,MAAM,CAAC;YAAC,IAAI,EAAE,MAAM,CAAA;SAAE,CAAC,CAAA;KAAE,CAAC;IAiE7E,+EAA+E;IAC/E,OAAO,CAAC,yBAAyB;IAcjC,iFAAiF;IACjF,OAAO,CAAC,oBAAoB;IA8B5B;;;;OAIG;YACW,eAAe;IAmD7B,OAAO,CAAC,cAAc;YAYR,YAAY;IAuK1B,cAAc,IAAI;QAAE,MAAM,EAAE,MAAM,CAAC;QAAC,OAAO,EAAE,MAAM,CAAC;QAAC,aAAa,EAAE,MAAM,CAAC;QAAC,OAAO,EAAE,iBAAiB,EAAE,CAAA;KAAE;IAI1G,gBAAgB,CAAC,IAAI,EAAE,MAAM,GAAG,aAAa;IAe7C,aAAa,CAAC,OAAO,CAAC,EAAE;QAAE,MAAM,CAAC,EAAE,QAAQ,GAAG,QAAQ,CAAA;KAAE,GAAG,OAAO,CAAC,iBAAiB,EAAE,CAAC;IAIvF,UAAU,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,iBAAiB,GAAG,SAAS,CAAC;IAIhE,UAAU,IAAI,iBAAiB,EAAE;IAIjC,SAAS,IAAI,aAAa;IAIpB,OAAO,CAAC,WAAW,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE;QAAE,KAAK,CAAC,EAAE,MAAM,CAAC;QAAC,MAAM,CAAC,EAAE,MAAM,CAAC;QAAC,IAAI,CAAC,EAAE,MAAM,CAAC;QAAC,EAAE,CAAC,EAAE,MAAM,CAAA;KAAE,GAAG,OAAO,CAAC,eAAe,EAAE,CAAC;IAInI,WAAW,CAAC,WAAW,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE;QAAE,IAAI,CAAC,EAAE,MAAM,CAAC;QAAC,EAAE,CAAC,EAAE,MAAM,CAAA;KAAE,GAAG,OAAO,CAAC,MAAM,CAAC;IAI3F,MAAM,CAAC,WAAW,EAAE,MAAM,EAAE,KAAK,EAAE,MAAM,GAAG,OAAO,CAAC,eAAe,GAAG,IAAI,CAAC;IAI3E,UAAU,CAAC,WAAW,EAAE,MAAM,GAAG,OAAO,CAAC,uBAAuB,EAAE,CAAC;IAMnE,YAAY,CAAC,WAAW,EAAE,MAAM,EAAE,QAAQ,EAAE,IAAI,CAAC,qBAAqB,EAAE,WAAW,CAAC,GAAG;QAAE,SAAS,CAAC,EAAE,MAAM,CAAA;KAAE,GAAG,OAAO,CAAC,IAAI,CAAC;IAQ7H,kBAAkB,CAAC,WAAW,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE;QAAE,UAAU,CAAC,EAAE,MAAM,CAAC;QAAC,IAAI,CAAC,EAAE,MAAM,CAAC;QAAC,EAAE,CAAC,EAAE,MAAM,CAAC;QAAC,KAAK,CAAC,EAAE,MAAM,CAAA;KAAE,GAAG,OAAO,CAAC,qBAAqB,EAAE,CAAC;IAIxJ,oBAAoB,CAAC,WAAW,EAAE,MAAM,GAAG,OAAO,CAAC,wBAAwB,GAAG,IAAI,CAAC;IAInF,oBAAoB,CAAC,WAAW,EAAE,MAAM,EAAE,MAAM,EAAE,wBAAwB,GAAG,OAAO,CAAC,IAAI,CAAC;IAI1F,uBAAuB,CAAC,WAAW,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,CAAC;IAI3D,QAAQ,IAAI,OAAO,CAAC,IAAI,CAAC;CA8BhC"}

package/dist/backend/process/engine.js

@@ -437,15 +437,13 @@ Instructions:
         try {
             const options = {
                 sessionId,
-                // Each process execution creates a fresh session, so claudeSessionId is always
-                // undefined on the first run. Subsequent runs of the same process still start
-                // fresh sessions. This is by design: process executions are isolated.
                 resumeSessionId: session.claudeSessionId,
                 userToken,
                 timeout,
                 serverUrl: this.serverUrl,
                 processName: definition.name,
                 runId: entry.runId,
+                model: definition.model,
             };
             if (files && files.length > 0) {
                 options.files = files;