feat: 2026-07-12~13 全量更新 - AI对话链路改造+H5 v4/v5+坐席端v5+上下文感知诊断+知识库迭代3

## H5 员工端 v4 (2026-07-13 00:48 已部署) - 人工按钮三态文案统一为"人工坐席" - 按钮位置移至发送键和语音按钮上方(垂直堆叠) - 点按钮直接调 store.shakeAgent()，删除 CallAgentModal 弹窗动画 - 截图快捷键提示改为"截图->粘贴：Alt+Shift+A-Ctrl+V ---> Ctrl+V" - 移动端隐藏截图提示(CSS 媒体查询) - AI转人工提示改为"已为您呼叫人工坐席，请稍等!" - 坐席接入提示改为"坐席正在查看您的信息，请等待处理回复！" - 删除"摇铃呼叫坐席"入口和文案 - 删除孤儿组件 MessageList.vue + shake 动画 CSS ## H5 员工端 v5 (2026-07-13 02:08 已部署) - RightPanel v2.1：删除"软件安装"和"资源权限"标签页 - 移除标签栏，智能推荐(DynamicRecommend)直接展示 - 删除 SoftwareDownloads/ApprovalLinks 引用和相关 CSS ## AI 对话链路全栈改造 Phase 1-6 (已部署) - Phase 1: Dify JSON输出 + 后端blocking解析 + 双WS推送 + 错误降级 - Phase 2: 关键词收窄(~25强意图词) + 两级分类Prompt + 删除前端checkApprovalIntent - Phase 3: WS扩展(ai_thinking+dynamic_recommend) + ai_structured气泡 + RightPanel v2 + 选项回传 - Phase 4: VisionService接入 + 图片消息融合(5秒窗口) + 降级策略 - Phase 5: 坐席端ai_thinking指示器 + ai_structured/byod_card渲染 + handleNewMessage修复 - Phase 6: diagnosis_stage(6值) + response_time_ms计时 + 慢响应告警(>10s) ## 坐席端 v5 (2026-07-13 01:38 已部署) - ai_structured/byod_card 只读渲染 - AI思考指示器 UI - handleNewMessage 透传 msg_type/extra_data 修复 - 布局优化v2.0: QuickReplyBar L1+L2悬浮 + ReplyBox左右分区 + 右栏260/560px切换 - 键盘快捷键v2.3: 纯数字路由 + ESC分层撤销 + Shift+Space用event.code ## 上下文感知智能诊断闭环 (2026-07-12 已部署) - 三层诊断(API→Script→AI) + 三段排队(VIP→info_locked→not locked) - 答题插队 + 五场景关闭 - 迁移052(6表+6列) + queue_service + quiz_service + closing_service - H5前端: QueueWaiting + RightPanel双Tab + InputBar三态 + ResolveConfirmCard - 坐席前端: pending_close结单流程 + 信息锁定(Dify步骤完成+有效回答率≥70%) ## 知识库迭代3 (2026-07-12 已部署) - 分诊交互(H5+坐席+Dify独立应用) - 拓扑预览(ECharts只读) - 代答排除(4种匹配器: keyword/regex/intent/category) - 迁移051 + 44文件43测试通过 ## 后端变更 - 6个Python文件改造(h5_ai_task.py/h5.py/ai_service.py/closing_service.py等) - funny_phrase_service.py: shake/connected/keyword 默认文案更新 - session_service.py: 企微消息文案同步 - 新增: queue.py/quiz.py/triage.py/exclusion_rules.py 等API端点 - 新增: diagnostic.py/quiz.py/triage_session.py 等模型 - 新增: closing_service/queue_service/quiz_service/triage_service 等服务 ## 文档更新 - CHANGELOG.md: 新增 [未发布] 区全部变更记录 - 项目管理主文档 v2.5: 新增v0.7.3版本 + 已完成看板 + 最近搞定 - 版本记录: 新增v0.7.3条目 - AI对话链路实施计划: Phase 1-6 全部标记✅已实施 - 新增架构图/时序图/类图(mermaid) ## 部署路径修正 - 服务器项目根路径: /opt/wecom-it-desk/ - 所有前端dist均为ro bind mount，只能在宿主机源路径操作 - 服务器nginx /h5/ 是静态文件服务(非proxy_pass) - elFinder上传二进制不可靠(MD5不匹配)，改用base64分块上传
2026-07-13 02:17:03 +08:00
parent bea288e414
commit 449c6d4875
176 changed files with 46637 additions and 4805 deletions
@@ -0,0 +1,184 @@
+# =============================================================================
+# 企微IT智能服务台 — 代答排除 Pydantic Schema
+# =============================================================================
+# 说明：代答排除模块的请求/响应数据模型，覆盖管理后台所有接口。
+# =============================================================================
+
+from datetime import datetime
+from typing import List, Optional
+
+from pydantic import BaseModel, Field
+
+
+# =============================================================================
+# 请求 Schema
+# =============================================================================
+
+class ExclusionRuleCreate(BaseModel):
+    """新建排除规则请求。
+
+    Attributes:
+        rule_name: 规则名称（唯一）
+        rule_description: 规则描述
+        priority: 优先级（P0/P1/P2/P3）
+        match_type: 匹配方式（keyword/regex/intent/category）
+        match_condition: 匹配条件
+        match_scope: 匹配范围
+        action_type: 命中后动作类型
+        transfer_message: 转人工提示语
+    """
+
+    rule_name: str = Field(..., min_length=1, max_length=200, description="规则名称")
+    rule_description: Optional[str] = Field(None, description="规则描述")
+    priority: str = Field("P2", description="优先级：P0/P1/P2/P3")
+    match_type: str = Field(..., description="匹配方式：keyword/regex/intent/category")
+    match_condition: str = Field(..., min_length=1, description="匹配条件")
+    match_scope: List[str] = Field(default_factory=lambda: ["ai_auto_reply"], description="匹配范围")
+    action_type: str = Field("transfer_human", description="命中后动作类型")
+    transfer_message: Optional[str] = Field(None, description="转人工提示语")
+
+
+class ExclusionRuleUpdate(BaseModel):
+    """编辑排除规则请求。
+
+    所有字段可选，仅更新传入的字段。
+
+    Attributes:
+        rule_name: 规则名称
+        rule_description: 规则描述
+        priority: 优先级
+        match_type: 匹配方式
+        match_condition: 匹配条件
+        match_scope: 匹配范围
+        action_type: 命中后动作类型
+        transfer_message: 转人工提示语
+    """
+
+    rule_name: Optional[str] = Field(None, min_length=1, max_length=200, description="规则名称")
+    rule_description: Optional[str] = Field(None, description="规则描述")
+    priority: Optional[str] = Field(None, description="优先级")
+    match_type: Optional[str] = Field(None, description="匹配方式")
+    match_condition: Optional[str] = Field(None, description="匹配条件")
+    match_scope: Optional[List[str]] = Field(None, description="匹配范围")
+    action_type: Optional[str] = Field(None, description="命中后动作类型")
+    transfer_message: Optional[str] = Field(None, description="转人工提示语")
+
+
+class ExclusionRuleToggle(BaseModel):
+    """启用/停用规则请求。
+
+    Attributes:
+        status: 目标状态（enabled/disabled）
+    """
+
+    status: str = Field(..., description="目标状态：enabled/disabled")
+
+
+class ExclusionTestRequest(BaseModel):
+    """测试匹配请求。
+
+    Attributes:
+        message: 测试消息文本
+        rule_id: 指定规则ID（可选，不指定则测试所有启用规则）
+    """
+
+    message: str = Field(..., min_length=1, description="测试消息文本")
+    rule_id: Optional[str] = Field(None, description="指定规则ID")
+
+
+# =============================================================================
+# 查询参数 Schema
+# =============================================================================
+
+class ExclusionRuleQuery(BaseModel):
+    """排除规则列表查询参数。
+
+    Attributes:
+        status: 状态筛选
+        match_type: 匹配方式筛选
+        priority: 优先级筛选
+        keyword: 关键词搜索（规则名称/描述）
+        page: 页码
+        page_size: 每页数量
+    """
+
+    status: Optional[str] = Field(None, description="状态筛选")
+    match_type: Optional[str] = Field(None, description="匹配方式筛选")
+    priority: Optional[str] = Field(None, description="优先级筛选")
+    keyword: Optional[str] = Field(None, description="关键词搜索")
+    page: int = Field(1, ge=1, description="页码")
+    page_size: int = Field(20, ge=1, le=100, description="每页数量")
+
+
+# =============================================================================
+# 响应 Schema
+# =============================================================================
+
+class ExclusionRuleResponse(BaseModel):
+    """排除规则响应。
+
+    Attributes:
+        id: 规则ID
+        rule_name: 规则名称
+        rule_description: 规则描述
+        priority: 优先级
+        match_type: 匹配方式
+        match_condition: 匹配条件
+        match_scope: 匹配范围
+        action_type: 命中后动作类型
+        transfer_message: 转人工提示语
+        status: 状态
+        hit_count: 命中次数
+        created_by: 创建人ID
+        created_at: 创建时间
+        updated_at: 更新时间
+    """
+
+    id: str = Field(..., description="规则ID")
+    rule_name: str = Field(..., description="规则名称")
+    rule_description: Optional[str] = Field(None, description="规则描述")
+    priority: str = Field(..., description="优先级")
+    match_type: str = Field(..., description="匹配方式")
+    match_condition: str = Field(..., description="匹配条件")
+    match_scope: List[str] = Field(default_factory=list, description="匹配范围")
+    action_type: str = Field(..., description="命中后动作类型")
+    transfer_message: Optional[str] = Field(None, description="转人工提示语")
+    status: str = Field(..., description="状态")
+    hit_count: int = Field(0, description="命中次数")
+    created_by: str = Field(..., description="创建人ID")
+    created_at: Optional[datetime] = Field(None, description="创建时间")
+    updated_at: Optional[datetime] = Field(None, description="更新时间")
+
+    model_config = {"from_attributes": True}
+
+
+class ExclusionTestResponse(BaseModel):
+    """测试匹配响应。
+
+    Attributes:
+        matched: 是否命中
+        matched_detail: 命中详情
+        rule_name: 命中的规则名称
+        action_type: 命中后动作类型
+    """
+
+    matched: bool = Field(False, description="是否命中")
+    matched_detail: Optional[str] = Field(None, description="命中详情")
+    rule_name: Optional[str] = Field(None, description="命中的规则名称")
+    action_type: Optional[str] = Field(None, description="命中后动作类型")
+
+
+class ExclusionStatsResponse(BaseModel):
+    """排除规则统计概要响应。
+
+    Attributes:
+        enabled_count: 启用规则数
+        disabled_count: 停用规则数
+        monthly_hits: 本月命中次数
+        monthly_transfers: 本月转人工次数
+    """
+
+    enabled_count: int = Field(0, description="启用规则数")
+    disabled_count: int = Field(0, description="停用规则数")
+    monthly_hits: int = Field(0, description="本月命中次数")
+    monthly_transfers: int = Field(0, description="本月转人工次数")
@@ -145,3 +145,48 @@ class TerminalBindingListResponse(BaseModel):

    list: List[TerminalBindingResponse] = Field(default_factory=list)
    total: int = 0
+
+
+# =============================================================================
+# 报修相关
+# =============================================================================
+
+class RepairRequest(BaseModel):
+    """提交报修请求。"""
+
+    terminal_sn: str = Field(..., min_length=1, max_length=64, description="终端序列号")
+    meetingroom_id: int = Field(..., description="企微会议室ID")
+    meetingroom_name: str = Field("", max_length=100, description="会议室名称")
+    device_type: str = Field(..., min_length=1, max_length=50, description="故障设备类型: projector/video_conf/aircon/desk_chair/network/other")
+    fault_description: str = Field(..., min_length=1, description="故障描述")
+    reporter_name: Optional[str] = Field(None, max_length=100, description="报修人姓名（为空则匿名）")
+    reporter_userid: Optional[str] = Field(None, max_length=64, description="报修人企微userid")
+
+
+class RepairResponse(BaseModel):
+    """报修提交响应。"""
+
+    repair_id: int = Field(..., description="报修记录ID")
+    conversation_id: str = Field(..., description="关联的IT工单会话ID")
+    status: int = Field(0, description="报修状态")
+
+
+# =============================================================================
+# 操作指南相关
+# =============================================================================
+
+class GuideItem(BaseModel):
+    """操作指南条目。"""
+
+    id: int = Field(..., description="指南ID")
+    category: str = Field(..., description="设备类型")
+    title: str = Field(..., description="指南标题")
+    brief: str = Field("", description="简要操作步骤")
+    detail_url: str = Field("", description="详细文档URL")
+    icon: str = Field("📋", description="图标emoji")
+
+
+class GuideListResponse(BaseModel):
+    """操作指南列表响应。"""
+
+    guides: List[GuideItem] = Field(default_factory=list, description="指南列表")
@@ -0,0 +1,324 @@
+# =============================================================================
+# 企微IT智能服务台 — 分诊交互 Pydantic Schema
+# =============================================================================
+# 说明：分诊模块的请求/响应数据模型，覆盖 H5 端和坐席端所有接口。
+# =============================================================================
+
+from datetime import datetime
+from typing import Any, Dict, List, Optional
+
+from pydantic import BaseModel, Field
+
+
+# =============================================================================
+# 基础嵌套模型
+# =============================================================================
+
+class TriageOption(BaseModel):
+    """分诊选项。
+
+    Attributes:
+        label: 选项标签文本
+        probability: AI 推荐概率（0.0-1.0）
+    """
+
+    label: str = Field(..., description="选项标签文本")
+    probability: Optional[float] = Field(None, ge=0.0, le=1.0, description="AI推荐概率")
+
+
+class TriageStep(BaseModel):
+    """分诊步骤。
+
+    Attributes:
+        question: 步骤问题文本
+        options: 选项列表
+    """
+
+    question: str = Field(..., description="步骤问题文本")
+    options: List[TriageOption] = Field(default_factory=list, description="选项列表")
+
+
+# =============================================================================
+# H5 端请求 Schema
+# =============================================================================
+
+class TriageStartRequest(BaseModel):
+    """发起分诊请求。
+
+    Attributes:
+        conversation_id: 会话ID
+        question: 员工问题文本
+    """
+
+    conversation_id: str = Field(..., description="会话ID")
+    question: str = Field(..., min_length=1, description="员工问题文本")
+
+
+class TriageStepRequest(BaseModel):
+    """提交步骤选择请求。
+
+    Attributes:
+        triage_id: 分诊会话ID
+        step_index: 当前步骤序号（0-based）
+        selected_label: 选择的选项标签
+    """
+
+    triage_id: str = Field(..., description="分诊会话ID")
+    step_index: int = Field(..., ge=0, description="当前步骤序号")
+    selected_label: str = Field(..., description="选择的选项标签")
+
+
+class TriageSkipRequest(BaseModel):
+    """跳过步骤请求。
+
+    Attributes:
+        triage_id: 分诊会话ID
+        step_index: 要跳过的步骤序号
+    """
+
+    triage_id: str = Field(..., description="分诊会话ID")
+    step_index: int = Field(..., ge=0, description="要跳过的步骤序号")
+
+
+class TriageTransferRequest(BaseModel):
+    """转人工请求。
+
+    Attributes:
+        triage_id: 分诊会话ID
+        context: 已收集的上下文列表
+    """
+
+    triage_id: str = Field(..., description="分诊会话ID")
+    context: List[str] = Field(default_factory=list, description="已收集的上下文列表")
+
+
+class TriageCompleteRequest(BaseModel):
+    """分诊完成请求。
+
+    Attributes:
+        triage_id: 分诊会话ID
+        context: 已收集的上下文列表
+    """
+
+    triage_id: str = Field(..., description="分诊会话ID")
+    context: List[str] = Field(default_factory=list, description="已收集的上下文列表")
+
+
+# =============================================================================
+# H5 端响应 Schema
+# =============================================================================
+
+class TriageStartResponse(BaseModel):
+    """发起分诊响应。
+
+    Attributes:
+        triage_id: 分诊会话ID
+        steps: 分诊步骤列表
+        total: 总步骤数
+        confidence: AI 置信度
+        urgency: 紧急度
+        suggested_route: AI 建议路由
+    """
+
+    triage_id: str = Field(..., description="分诊会话ID")
+    steps: List[TriageStep] = Field(default_factory=list, description="分诊步骤列表")
+    total: int = Field(0, description="总步骤数")
+    confidence: Optional[float] = Field(None, description="AI 置信度")
+    urgency: str = Field("medium", description="紧急度")
+    suggested_route: Optional[str] = Field(None, description="AI 建议路由")
+
+
+class TriageStepResponse(BaseModel):
+    """提交步骤选择响应。
+
+    Attributes:
+        next_step: 下一步骤数据（无下一步时为 null）
+        collected_context: 已收集的上下文列表
+    """
+
+    next_step: Optional[TriageStep] = Field(None, description="下一步骤数据")
+    collected_context: List[str] = Field(default_factory=list, description="已收集的上下文列表")
+
+
+class TriageTransferResponse(BaseModel):
+    """转人工响应。
+
+    Attributes:
+        conversation_id: 会话ID
+        status: 会话状态
+    """
+
+    conversation_id: str = Field(..., description="会话ID")
+    status: str = Field("waiting_agent", description="会话状态")
+
+
+class TriageCompleteResponse(BaseModel):
+    """分诊完成响应。
+
+    Attributes:
+        reply: AI 生成的最终回复
+        confidence: AI 置信度
+    """
+
+    reply: str = Field(..., description="AI 生成的最终回复")
+    confidence: float = Field(0.0, description="AI 置信度")
+
+
+# =============================================================================
+# 坐席端请求 Schema
+# =============================================================================
+
+class TriageRouteRequest(BaseModel):
+    """坐席路由操作请求。
+
+    Attributes:
+        route_action: 路由动作（ai_self/human/auto_approval/skip）
+        route_note: 路由备注
+    """
+
+    route_action: str = Field(..., description="路由动作：ai_self/human/auto_approval/skip")
+    route_note: Optional[str] = Field(None, description="路由备注")
+
+
+class TriageExcludeOptionsRequest(BaseModel):
+    """坐席排除/推荐分诊选项请求。
+
+    Attributes:
+        excluded_labels: 要排除的选项标签列表
+        recommended_label: 推荐的选项标签
+    """
+
+    excluded_labels: List[str] = Field(default_factory=list, description="要排除的选项标签列表")
+    recommended_label: Optional[str] = Field(None, description="推荐的选项标签")
+
+
+# =============================================================================
+# 坐席端响应 Schema
+# =============================================================================
+
+class TriageSessionResponse(BaseModel):
+    """分诊会话列表项响应。
+
+    Attributes:
+        id: 分诊会话ID
+        conversation_id: 会话ID
+        user_id: 员工ID
+        user_name: 员工姓名
+        user_dept: 员工部门
+        request_title: 问题标题
+        problem_type: 问题类型
+        problem_category: 问题分类
+        confidence: AI 置信度
+        urgency: 紧急度
+        suggested_route: AI 建议路由
+        status: 分诊状态
+        route_action: 路由动作
+        route_note: 路由备注
+        operator_id: 操作坐席ID
+        created_at: 创建时间
+        operated_at: 操作时间
+    """
+
+    id: str = Field(..., description="分诊会话ID")
+    conversation_id: str = Field(..., description="会话ID")
+    user_id: str = Field(..., description="员工ID")
+    user_name: Optional[str] = Field(None, description="员工姓名")
+    user_dept: Optional[str] = Field(None, description="员工部门")
+    request_title: str = Field(..., description="问题标题")
+    problem_type: Optional[str] = Field(None, description="问题类型")
+    problem_category: Optional[str] = Field(None, description="问题分类")
+    confidence: Optional[float] = Field(None, description="AI 置信度")
+    urgency: str = Field("medium", description="紧急度")
+    suggested_route: Optional[str] = Field(None, description="AI 建议路由")
+    status: str = Field("pending", description="分诊状态")
+    route_action: Optional[str] = Field(None, description="路由动作")
+    route_note: Optional[str] = Field(None, description="路由备注")
+    operator_id: Optional[str] = Field(None, description="操作坐席ID")
+    created_at: Optional[datetime] = Field(None, description="创建时间")
+    operated_at: Optional[datetime] = Field(None, description="操作时间")
+
+    model_config = {"from_attributes": True}
+
+
+class TriageDetailResponse(BaseModel):
+    """分诊详情响应（含完整数据）。
+
+    Attributes:
+        id: 分诊会话ID
+        conversation_id: 会话ID
+        user_id: 员工ID
+        user_name: 员工姓名
+        user_dept: 员工部门
+        user_level: 员工IT技能等级
+        device_info: 设备信息
+        request_title: 问题标题
+        request_content: 问题原文
+        source: 来源渠道
+        problem_type: 问题类型
+        problem_category: 问题分类
+        confidence: AI 置信度
+        urgency: 紧急度
+        suggested_route: AI 建议路由
+        matched_knowledge: 匹配到的知识条目
+        match_score: 知识匹配分数
+        context_tags: 上下文标签列表
+        triage_steps: 分诊步骤数据
+        collected_context: 已收集的上下文列表
+        status: 分诊状态
+        route_action: 路由动作
+        route_note: 路由备注
+        operator_id: 操作坐席ID
+        created_at: 创建时间
+        updated_at: 更新时间
+        operated_at: 操作时间
+    """
+
+    id: str = Field(..., description="分诊会话ID")
+    conversation_id: str = Field(..., description="会话ID")
+    user_id: str = Field(..., description="员工ID")
+    user_name: Optional[str] = Field(None, description="员工姓名")
+    user_dept: Optional[str] = Field(None, description="员工部门")
+    user_level: Optional[str] = Field(None, description="员工IT技能等级")
+    device_info: Optional[str] = Field(None, description="设备信息")
+    request_title: str = Field(..., description="问题标题")
+    request_content: str = Field(..., description="问题原文")
+    source: str = Field("wecom_h5", description="来源渠道")
+    problem_type: Optional[str] = Field(None, description="问题类型")
+    problem_category: Optional[str] = Field(None, description="问题分类")
+    confidence: Optional[float] = Field(None, description="AI 置信度")
+    urgency: str = Field("medium", description="紧急度")
+    suggested_route: Optional[str] = Field(None, description="AI 建议路由")
+    matched_knowledge: Optional[str] = Field(None, description="匹配到的知识条目")
+    match_score: Optional[float] = Field(None, description="知识匹配分数")
+    context_tags: List[str] = Field(default_factory=list, description="上下文标签列表")
+    triage_steps: List[Dict[str, Any]] = Field(default_factory=list, description="分诊步骤数据")
+    collected_context: List[str] = Field(default_factory=list, description="已收集的上下文列表")
+    status: str = Field("pending", description="分诊状态")
+    route_action: Optional[str] = Field(None, description="路由动作")
+    route_note: Optional[str] = Field(None, description="路由备注")
+    operator_id: Optional[str] = Field(None, description="操作坐席ID")
+    created_at: Optional[datetime] = Field(None, description="创建时间")
+    updated_at: Optional[datetime] = Field(None, description="更新时间")
+    operated_at: Optional[datetime] = Field(None, description="操作时间")
+
+    model_config = {"from_attributes": True}
+
+
+class TriageStatsResponse(BaseModel):
+    """分诊看板统计概要响应。
+
+    Attributes:
+        pending_total: 待分诊总数
+        today_triaged: 今日已分诊数
+        ai_self_count: AI 自答数
+        human_count: 转人工数
+        auto_approval_count: 自动审批数
+        avg_duration_sec: 平均耗时（秒）
+    """
+
+    pending_total: int = Field(0, description="待分诊总数")
+    today_triaged: int = Field(0, description="今日已分诊数")
+    ai_self_count: int = Field(0, description="AI 自答数")
+    human_count: int = Field(0, description="转人工数")
+    auto_approval_count: int = Field(0, description="自动审批数")
+    avg_duration_sec: float = Field(0.0, description="平均耗时（秒）")