chore: sync pending project changes

release: prepare v3.1.0
docs: align README badges in one row
2026-09-13 00:14:59 +00:00 · 2026-08-04 15:23:19 +08:00 · 2026-08-04 15:18:57 +08:00 · 2026-07-30 00:22:22 +08:00 · 2026-07-30 00:15:44 +08:00
50 changed files with 1342 additions and 1602 deletions
--- a/.github/workflows/publish-images.yml
+++ b/.github/workflows/publish-images.yml
@@ -0,0 +1,107 @@
+name: Publish Docker Images
+
+on:
+  push:
+    tags:
+      - 'v*.*.*'
+
+concurrency:
+  group: publish-images-${{ github.ref_name }}
+  cancel-in-progress: false
+
+env:
+  REGISTRY: ghcr.io
+  IMAGE_OWNER: ${{ github.repository_owner }}
+  RELEASE_TAG: ${{ github.ref_name }}
+
+jobs:
+  publish:
+    name: Publish ${{ matrix.image }}
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      packages: write
+      attestations: write
+      id-token: write
+
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          - image: ruoyi-ai-backend
+            source: local
+            context: .
+            dockerfile: docs/docker/ruoyi-ai/Dockerfile.backend
+          - image: ruoyi-ai-mysql
+            source: local
+            context: .
+            dockerfile: docs/docker/ruoyi-ai/Dockerfile.mysql
+          - image: ruoyi-ai-admin
+            source: admin
+            context: build/ruoyi-admin
+            dockerfile: build/ruoyi-admin/apps/web-antd/Dockerfile
+          - image: ruoyi-ai-web
+            source: web
+            context: build/ruoyi-web
+            dockerfile: build/ruoyi-web/Dockerfile.frontend
+
+    steps:
+      - name: Checkout backend repository
+        uses: actions/checkout@v6
+        with:
+          ref: ${{ env.RELEASE_TAG }}
+          fetch-depth: 1
+
+      - name: Checkout admin repository
+        if: matrix.source == 'admin'
+        uses: actions/checkout@v6
+        with:
+          repository: ${{ github.repository_owner }}/ruoyi-admin
+          ref: ${{ env.RELEASE_TAG }}
+          path: build/ruoyi-admin
+          fetch-depth: 1
+
+      - name: Checkout web repository
+        if: matrix.source == 'web'
+        uses: actions/checkout@v6
+        with:
+          repository: ${{ github.repository_owner }}/ruoyi-web
+          ref: ${{ env.RELEASE_TAG }}
+          path: build/ruoyi-web
+          fetch-depth: 1
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Log in to GHCR
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.REGISTRY }}
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Extract Docker metadata
+        id: meta
+        uses: docker/metadata-action@v5
+        with:
+          images: ${{ env.REGISTRY }}/${{ env.IMAGE_OWNER }}/${{ matrix.image }}
+          tags: |
+            type=raw,value=${{ env.RELEASE_TAG }}
+            type=raw,value=latest
+          labels: |
+            org.opencontainers.image.source=https://github.com/${{ github.repository }}
+            org.opencontainers.image.version=${{ env.RELEASE_TAG }}
+
+      - name: Build and push image
+        uses: docker/build-push-action@v6
+        with:
+          context: ${{ matrix.context }}
+          file: ${{ matrix.dockerfile }}
+          platforms: linux/amd64
+          push: true
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.outputs.labels }}
+          cache-from: type=gha,scope=${{ matrix.image }}
+          cache-to: type=gha,mode=max,scope=${{ matrix.image }}
+          provenance: true
+          sbom: true
--- a/.gitignore
+++ b/.gitignore
@@ -23,6 +23,9 @@ target/
 .idea
 .claude
 .github
+!.github/
+!.github/workflows/
+!.github/workflows/**
 *.iws
 *.iml
 *.ipr
--- a/FASTJSON_REMOVAL_REPORT.md
+++ b/FASTJSON_REMOVAL_REPORT.md
@@ -1,124 +0,0 @@
-# FastJson 安全漏洞修复报告
-
-## 修复概述
- **修复日期**: 2026-07-29
- **漏洞等级**: 🔴 Critical (严重)
- **修复状态**: ✅ 已完成
-
-## 漏洞描述
-项目使用的 FastJson 1.2.83 版本存在严重的反序列化远程代码执行(RCE)漏洞，包括：
- CVE-2022-25845
- CVE-2023-21931
- 多个未公开的反序列化漏洞
-
-攻击者可通过构造恶意JSON实现远程代码执行，具有极高的安全风险。
-
-## 修复方案
-完全移除 FastJson 依赖，替换为 Spring Boot 内置的 Jackson 库。
-
-## 修复详情
-
-### 1. POM 依赖修改
-
-#### 根 pom.xml
- 删除: fastjson.version 属性定义
- 删除: fastjson 依赖声明
- 状态: ✅ 已完成
-
-#### ruoyi-common-chat/pom.xml
- 删除: fastjson 依赖
- 新增: jackson-databind 依赖
- 状态: ✅ 已完成
-
-### 2. Java 代码修改
-
-共修改了 **6个Java文件**，替换所有 FastJson API 为 Jackson API。
-
-#### 修改文件列表:
-1. ✅ QwenFileUploadUtils.java - 千问文件上传工具
-2. ✅ ChatRequest.java - 聊天请求对象
-3. ✅ MailSendNode.java - 邮件发送节点
-4. ✅ SwitcherNode.java - 条件分支节点
-5. ✅ AbstractAuthWeChatEnterpriseRequest.java - 企业微信登录
-6. ✅ AuthDingTalkV2Request.java - 钉钉登录
-
-### 3. API 替换对照表
-
-| 操作 | FastJson | Jackson |
-|------|----------|---------|
-| 解析JSON | JSONObject.parseObject(str) | objectMapper.readTree(str) |
-| 获取字符串 | json.getString("key") | json.get("key").asText() |
-| 获取整数 | json.getIntValue("key") | json.get("key").asInt() |
-| 判断包含 | json.containsKey("key") | json.has("key") |
-| 对象转JSON | JSON.toJSONString(obj) | objectMapper.writeValueAsString(obj) |
-
-## 特殊说明 - JustAuth库兼容
-
-由于第三方 JustAuth 库的 AuthUser.rawUserInfo 字段需要 FastJson 的 JSONObject 类型，
-在两个社交登录文件中保留了最小化的 FastJson 使用：
-
- AbstractAuthWeChatEnterpriseRequest.java
- AuthDingTalkV2Request.java
-
-**使用方式**: 仅用于格式转换（Jackson JsonNode → FastJson JSONObject）
-**安全性**: ✅ 不涉及反序列化，仅数据转换，安全可控
-
-## 验证结果
-
-### 编译验证
-```
-mvn clean compile -DskipTests
-```
-**结果**: ✅ BUILD SUCCESS (所有38个模块编译通过)
-**耗时**: 01:26 min
-
-### 代码检查
- FastJson 导入残留: 0个（除兼容性转换）
- POM 依赖残留: 0个
-
-## 安全提升对比
-
-### 修复前
- ❌ FastJson 1.2.83 (严重RCE漏洞)
- ❌ 全局攻击面暴露
- ❌ 可被恶意JSON远程执行代码
-
-### 修复后  
- ✅ Jackson 2.18.2 (Spring Boot内置，安全稳定)
- ✅ 移除反序列化RCE攻击面
- ✅ 显著提升系统安全性
- ⚠️ 保留最小化FastJson使用（仅格式转换）
-
-## 受影响的功能模块
-
-1. ✅ 千问文件上传
-2. ✅ 聊天请求处理
-3. ✅ 工作流邮件发送
-4. ✅ 工作流条件分支
-5. ✅ 企业微信登录
-6. ✅ 钉钉登录
-
-**测试建议**: 重点测试以上功能模块的JSON处理和社交登录功能
-
-## 后续优化建议
-
-1. **监控JustAuth更新**: 等待其支持Jackson后完全移除FastJson
-2. **功能测试**: 进行完整的回归测试
-3. **安全监控**: 关注Jackson的安全更新
-
-## 总结
-
-✅ **修复完成度**: 95%
- 主要业务代码: 100% 完成
- 第三方库兼容: 保留最小化使用
-
-🎯 **安全成果**:
- 消除了 FastJson 1.2.83 的严重RCE漏洞
- 提升了整体系统安全防护能力
- 所有修改已通过编译验证
-
---
-
-**修复人员**: Claude Code AI  
-**审核状态**: ✅ 待人工审核  
-**建议操作**: 合并前进行完整功能测试
--- a/RAG_TEST_REPORT_2026-07-20.md
+++ b/RAG_TEST_REPORT_2026-07-20.md
@@ -1,64 +0,0 @@
-# RAG 完整修复与全量验收报告
-
-验收时间：2026-07-21（Asia/Shanghai）  
-验收对象：当前未提交工作区（保留原有改动）
-
-## 结论
-
-计划内的 1–15 项工程缺陷已完成代码修复，默认 Maven 构建已从“跳过测试”改为真实执行测试。全仓 37 个 reactor 模块测试成功，`ruoyi-chat` 49/49 通过，两个前端生产构建通过，`git diff --check` 通过。
-
-本机已运行 MySQL、Redis、MinIO 和 Weaviate 1.30.0；Milvus/Qdrant 容器以及有效的 embedding/chat/rerank provider 凭证不存在，因此这三项真实 provider/存储引擎冒烟被标记为环境限制，不影响确定性代码验收。
-
-## 1–15 项验收
-
-| # | 状态 | 修复/证据 |
-|---|---|---|
-| 1 | 通过 | Markdown/Java/字符分片的边界、空文档、超长块回归通过。 |
-| 2 | 通过 | Supervisor 每轮仅保留一个 RAG 入口，不再用已含 RAG 的 prompt 重复检索。 |
-| 3 | 通过 | 历史消息进入 Supervisor prompt，检索 query 与最终 prompt 分离。 |
-| 4 | 通过 | `fid` 稳定 ID 贯穿 DB/三种向量库/RRF，融合去重回归通过。 |
-| 5 | 通过 | aiflow vector/hybrid 复用统一检索服务；graph 明确返回不支持，不再伪装为 vector。 |
-| 6 | 通过 | 重解析改为先写新 fid、再清旧向量、最后替换 DB；失败补偿新向量；删片段/附件/库遇向量删除失败即中止。 |
-| 7 | 通过 | embedding/rerank provider 使用 prototype 实例，工厂缓存可按模型刷新，避免跨配置污染。 |
-| 8 | 通过 | `similarityThreshold` 仅用于粗召回；`rerankScoreThreshold` 仅在 rerank 真实成功后生效，回归测试通过。 |
-| 9 | 通过 | 默认配置和 Compose 统一为 Weaviate 1.30.0、`28080:8080`。 |
-| 10 | 通过 | 三种策略均使用 `embedAll`；Weaviate batch objects、Milvus `addAll`、Qdrant `addAll`。 |
-| 11 | 通过 | upload/parse/retrieval 权限保留，parse/retrieval 增加分布式防重复提交，upload 由现有知识库+文件名唯一约束兜底。 |
-| 12 | 通过 | 分隔符使用字面量语义，`|`/`.`/`*` 回归通过。 |
-| 13 | 通过 | hybrid 通道失败可降级到 vector；所有可用通道都失败时抛出明确业务异常。 |
-| 14 | 通过 | Weaviate client 稳定懒加载单例；schema 仅在已存在或创建成功后进入缓存。 |
-| 15 | 通过 | 工厂新增严格 `getStrategy(type)`，知识库 `vectorModel` 优先，空值才回退全局，非法值直接报错。 |
-
-## 其他完成项
-
- 多知识库并行检索，按 `kid + docId + fid` 去重，统一上限和字符预算。
- 5 分钟短 TTL 检索缓存，key 覆盖检索参数，知识数据变更主动失效。
- rerank 仅保留 provider 实际返回的文档。
- 知识库文档数改为 group-by 查询，消除该 N+1。
- Milvus/Qdrant/Weaviate 的删 collection/doc/fid 语义对齐；Milvus 删库改为 drop collection。
- MCP `npx` 根据操作系统解析，支持系统属性/环境变量覆盖。
- `fid` 非空唯一、`doc_id varchar(32)`、租户/用户索引与可重复执行迁移脚本已提供。
- 用户端聊天页已接入知识库列表和最小选择器。
-
-## 测试记录
-
-| 检查 | 结果 |
-|---|---|
-| `mvn -Pdev test` | 37/37 reactor 模块 SUCCESS；`ruoyi-chat` 49/49 |
-| `mvn -Pdev -pl ruoyi-modules/ruoyi-aiflow -am -DskipTests compile` | 21/21 SUCCESS |
-| `ruoyi-web: pnpm build` | SUCCESS，2621 modules transformed |
-| `ruoyi-admin: pnpm build` | SUCCESS，10/10 build tasks |
-| `git diff --check` | SUCCESS，无空白错误 |
-| Weaviate/MySQL/Redis/MinIO | Docker 服务运行，Weaviate 1.30.0 映射 28080 |
-| 三向量库 Docker 集成 | SUCCESS；Weaviate 1.30.0、Milvus 2.5.7、Qdrant 1.17.0 真实写入/检索/删除测试 3/3 通过 |
-| 真实 embedding/chat/rerank | 环境限制：当前配置为无效/占位凭证 |
-
-本轮未创建新的 `codex_rag_verify_` 持久化数据；上一轮验收数据已清理，未动现有非测试数据。
-
-## 2026-07-21 三向量库 Docker 补充验收
-
- 启动并保留 `ruoyi-rag-milvus`、`ruoyi-rag-milvus-etcd`、`ruoyi-rag-milvus-minio`、`ruoyi-rag-qdrant`，四个容器健康检查均为 `healthy`。
- Milvus 专用 MinIO 仅在 Docker 内网可达，没有占用宿主机 9000/9001；Milvus 映射 19530/9091，Qdrant 映射 6333/6334。
- `ThreeVectorStoresDockerIT` 使用 32 维确定性 embedding，对三库逐一验证 batch write、vector search、fid delete、docId delete 和 drop collection，3/3 通过。
- 首轮测试发现 Milvus `autoFlush=false` 导致批量写入后不可立即检索、元数据删除不可立即见；改为写入和删除返回前 flush 后通过。
- 清理后 Weaviate/Qdrant 的 `CodexRagVerify*` collection 计数均为 0，Milvus collection 也由测试 finally 成功 drop；本轮未写入 MySQL 或 OSS 测试数据。
--- a/README.md
+++ b/README.md
@@ -2,11 +2,7 @@

 <div align="center">

-[![Contributors][contributors-shield]][contributors-url]
-[![Forks][forks-shield]][forks-url]
-[![Stargazers][stars-shield]][stars-url]
-[![Issues][issues-shield]][issues-url]
-[![MIT License][license-shield]][license-url]
+[![Contributors][contributors-shield]][contributors-url] [![Forks][forks-shield]][forks-url] [![Stargazers][stars-shield]][stars-url] [![Issues][issues-shield]][issues-url] [![MIT License][license-shield]][license-url]


 <p align="center">
@@ -17,236 +13,235 @@

 <img src="docs/image/logo.png" alt="RuoYi AI Logo" width="120" height="120">

-### 企业级AI助手平台
+### Enterprise-Grade AI Assistant Platform

-*开箱即用的全栈AI平台，支持多智能体协同、Supervisor模式编排、多种决策模式、RAG技术和流程编排能力*
+*An out-of-the-box full-stack AI platform supporting multi-agent collaboration, Supervisor mode orchestration, and multiple decision models, with advanced RAG technology and visual workflow orchestration capabilities*

-**[English](README_EN.md)** | **[📖 使用文档](https://doc.ruoyiai.chat/)** |
-**[🚀 在线体验](https://web.ruoyiai.chat/)** | **[🐛 问题反馈](https://github.com/ageerle/ruoyi-ai/issues)** | **[💡 功能建议](https://github.com/ageerle/ruoyi-ai/issues)**
+**[中文](README_ZH.md)** | **[📖 Documentation](https://doc.ruoyiai.chat/)** |
+**[🚀 Live Demo](https://web.ruoyiai.chat/)** | **[🐛 Report Issues](https://github.com/ageerle/ruoyi-ai/issues)** | **[💡 Feature Requests](https://github.com/ageerle/ruoyi-ai/issues)**

 </div>


-## ✨ 核心亮点
-
-|    模块     | 现有能力 
-|:---------:|---
-| **模型管理**  | 多模型接入(DeepSeek/智谱/MIMO/百炼/OpenAI)、多模态理解、Coze/DIFY/FastGPT/RAGFlow平台集成
-| **知识管理**  | 本地RAG + 向量库(Milvus/Weaviate/Qdrant)  + 文档解析
-| **工具管理**  | Mcp协议集成、Skills能力 + 可扩展工具生态                             
-| **流程编排**  | 可视化工作流设计器、节点拖拽编排、SSE流式执行,目前已经支持模型调用,邮件发送,人工审核等节点  
-| **智能体管理** | 基于Langchain4j的Agent框架、Supervisor模式编排,支持多种决策模型,可以灵活搭配工具,skills          


-### 项目源码
+## ✨ Core Features

-| 项目模块     | GitHub 仓库                                             | Gitee 仓库                                             | GitCode 仓库                                             |
+| Module | Current Capabilities |
+|:---:|---|
+| **Model Management** | Multi-model integration (DeepSeek/Zhipu/MIMO/Bailian/OpenAI), multi-modal understanding, Coze/DIFY/FastGPT/RAGFlow platform integration |
+| **Knowledge Management** | Local RAG + Vector DB (Milvus/Weaviate/Qdrant) + Document parsing |
+| **Tool Management** | MCP protocol integration, Skills capability + Extensible tool ecosystem |
+| **Workflow Orchestration** | Visual workflow designer, drag-and-drop node orchestration, SSE streaming execution, currently supports model calls, email sending, manual review, and other nodes |
+| **Multi-Agent** | Agent framework based on Langchain4j, Supervisor mode orchestration, supports multiple decision models, can flexibly combine tools and skills |
+
+### Project Repositories
+
+| Module     | GitHub Repository                                             | Gitee Repository                                             | GitCode Repository                                             |
 |----------|-------------------------------------------------------|------------------------------------------------------|--------------------------------------------------------|
-| 🔧 后端服务  | [ruoyi-ai](https://github.com/ageerle/ruoyi-ai)       | [ruoyi-ai](https://gitee.com/ageerle/ruoyi-ai)       | [ruoyi-ai](https://gitcode.com/ageerle/ruoyi-ai)       |
-| 🎨 用户前端  | [ruoyi-web](https://github.com/ageerle/ruoyi-web)     | [ruoyi-web](https://gitee.com/ageerle/ruoyi-web)     | [ruoyi-web](https://gitcode.com/ageerle/ruoyi-web)     |
-| 🛠️ 管理后台 | [ruoyi-admin](https://github.com/ageerle/ruoyi-admin) | [ruoyi-admin](https://gitee.com/ageerle/ruoyi-admin) | [ruoyi-admin](https://gitcode.com/ageerle/ruoyi-admin) |
-| 🎬 短剧平台  | [ruoyi-drama](https://github.com/ageerle/ruoyi-drama) | [ruoyi-drama](https://gitee.com/ageerle/ruoyi-drama) | [ruoyi-drama](https://gitcode.com/ageerle/ruoyi-drama)   |
-| 🤖 编程助手  | [ruoyi-copilot](https://github.com/ageerle/ruoyi-copilot) | [ruoyi-copilot](https://gitee.com/ageerle/ruoyi-copilot) | [ruoyi-copilot](https://gitcode.com/ageerle/ruoyi-copilot) |
-| 📱 小程序端  | [ruoyi-uniapp](https://github.com/ageerle/ruoyi-uniapp) | [ruoyi-uniapp](https://gitee.com/ageerle/ruoyi-uniapp) | [ruoyi-uniapp](https://gitcode.com/ageerle/ruoyi-uniapp) |
+| 🔧 Backend  | [ruoyi-ai](https://github.com/ageerle/ruoyi-ai)       | [ruoyi-ai](https://gitee.com/ageerle/ruoyi-ai)       | [ruoyi-ai](https://gitcode.com/ageerle/ruoyi-ai)       |
+| 🎨 User Frontend  | [ruoyi-web](https://github.com/ageerle/ruoyi-web)     | [ruoyi-web](https://gitee.com/ageerle/ruoyi-web)     | [ruoyi-web](https://gitcode.com/ageerle/ruoyi-web)     |
+| 🛠️ Admin Panel | [ruoyi-admin](https://github.com/ageerle/ruoyi-admin) | [ruoyi-admin](https://gitee.com/ageerle/ruoyi-admin) | [ruoyi-admin](https://gitcode.com/ageerle/ruoyi-admin) |
+| 🎬 Drama | [ruoyi-drama](https://github.com/ageerle/ruoyi-drama) | [ruoyi-drama](https://gitee.com/ageerle/ruoyi-drama) | [ruoyi-drama](https://gitcode.com/ageerle/ruoyi-drama) |
+| 🤖 Copilot | [ruoyi-copilot](https://github.com/ageerle/ruoyi-copilot) | [ruoyi-copilot](https://gitee.com/ageerle/ruoyi-copilot) | [ruoyi-copilot](https://gitcode.com/ageerle/ruoyi-copilot) |
+| 📱 Mini-App | [ruoyi-uniapp](https://github.com/ageerle/ruoyi-uniapp) | [ruoyi-uniapp](https://gitee.com/ageerle/ruoyi-uniapp) | [ruoyi-uniapp](https://gitcode.com/ageerle/ruoyi-uniapp) |

-### 合作项目
-| 项目名称           | GitHub 仓库                                             | Gitee 仓库                                             
+### Partner Projects
+| Project Name           | GitHub Repository                                             | Gitee Repository |
 |----------------|-------------------------------------------------------|------------------------------------------------------|
 | element-plus-x | [element-plus-x](https://github.com/element-plus-x/Element-Plus-X)       | [element-plus-x](https://gitee.com/he-jiayue/element-plus-x)       |

-## 🛠️ 技术架构
+## 🛠️ Technical Architecture

-### 核心框架
- **后端架构**：Spring Boot 3.5.8 + Langchain4j
- **数据存储**：MySQL 8.0 + Redis + 向量数据库（Milvus/Weaviate/Qdrant）
- **前端技术**：Vue 3 + Vben Admin + element-plus-x
- **安全认证**：Sa-Token + JWT 双重保障
- **文档处理**：PDF、Word、Excel 解析，图像智能分析
- **实时通信**：WebSocket 实时通信，SSE 流式响应
- **系统监控**：完善的日志体系、性能监控、服务健康检查
+### Core Framework
+- **Backend**: Spring Boot 3.5.8 + Langchain4j
+- **Data Storage**: MySQL 8.0 + Redis + Vector Databases (Milvus/Weaviate/Qdrant)
+- **Frontend**: Vue 3 + Vben Admin + element-plus-x
+- **Security**: Sa-Token + JWT dual-layer security
+- **Document Processing**: PDF, Word, Excel parsing, intelligent image analysis
+- **Real-time Communication**: WebSocket real-time communication, SSE streaming response
+- **System Monitoring**: Comprehensive logging system, performance monitoring, service health checks

-## 🐳 Docker 部署
+## 🐳 Docker Deployment

-本项目提供两种 Docker 部署方式：
+This project provides two Docker deployment methods:

-### 方式一：一键启动所有服务（推荐）
+### Method 1: One-click Start All Services (Recommended)

-使用 `docker-compose-all.yaml` 可以一键启动所有服务（包括后端、管理端、用户端及依赖服务）：
+Use `docker-compose-all.yaml` to start all services at once (including backend, admin panel, user frontend, and dependencies):

 ```bash
-# 克隆仓库
+# Clone the repository
 git clone https://github.com/ageerle/ruoyi-ai.git
 cd ruoyi-ai

-# 启动所有服务（从镜像仓库拉取预构建镜像）
-docker-compose -f docker-compose-all.yaml up -d
+# Start all services (pull pre-built images from GHCR)
+docker compose -f docs/docker/ruoyi-ai/docker-compose-all.yaml up -d

-# 查看服务状态
-docker-compose -f docker-compose-all.yaml ps
+# Check service status
+docker compose -f docs/docker/ruoyi-ai/docker-compose-all.yaml ps

-# 访问服务
-# 管理端: http://localhost:25666 (admin / admin123)
-# 用户端: http://localhost:25137
-# 后端API: http://localhost:26039
+# Access services
+# Admin Panel: http://localhost:25666 (admin / admin123)
+# User Frontend: http://localhost:25137
+# Backend API: http://localhost:26039
 ```

-### 方式二：分步部署（源码编译）
+### Method 2: Step-by-step Deployment (Source Build)

-如果您需要从源码构建后端服务，请按照以下步骤操作：
+If you need to build backend services from source, follow these steps:

-#### 第一步：部署后端服务
+#### Step 1: Deploy Backend Service

 ```bash
-# 进入后端项目目录
+# Enter backend project directory
 cd ruoyi-ai

-# 启动后端服务（源码编译构建）
+# Start backend service (build from source)
 docker-compose up -d --build

-# 等待后端服务启动完成
+# Wait for backend service to start
 docker-compose logs -f backend
 ```

-#### 第二步：部署管理端
+#### Step 2: Deploy Admin Panel

 ```bash
-# 进入管理端项目目录
+# Enter admin panel project directory
 cd ruoyi-admin

-# 构建并启动管理端
+# Build and start admin panel
 docker-compose up -d --build

-# 访问管理端
-# 地址: http://localhost:5666
+# Access admin panel
+# URL: http://localhost:5666
 ```

-#### 第三步：部署用户端（可选）
+#### Step 3: Deploy User Frontend (Optional)

 ```bash
-# 进入用户端项目目录
+# Enter user frontend project directory
 cd ruoyi-web

-# 构建并启动用户端
+# Build and start user frontend
 docker-compose up -d --build

-# 访问用户端
-# 地址: http://localhost:5137
+# Access user frontend
+# URL: http://localhost:5137
 ```

-### 服务端口说明
+### Service Ports

-| 服务 | 一键启动端口 | 分步部署端口 | 说明 |
+| Service | One-click Port | Step-by-step Port | Description |
 |------|-------------|-------------|------|
-| 管理端 | 25666 | 5666 | 管理后台访问地址 |
-| 用户端 | 25137 | 5137 | 用户前端访问地址 |
-| 后端服务 | 26039 | 6039 | 后端 API 服务 |
-| MySQL | 23306 | 23306 | 数据库服务 |
-| Redis | 26379 | 6379 | 缓存服务 |
-| Weaviate | 28080 | 28080 | 向量数据库 |
-| MinIO API | 29000 | 9000 | 对象存储 API |
-| MinIO Console | 29090 | 9090 | 对象存储控制台 |
+| Admin Panel | 25666 | 5666 | Admin backend access |
+| User Frontend | 25137 | 5137 | User frontend access |
+| Backend Service | 26039 | 6039 | Backend API service |
+| MySQL | 23306 | 23306 | Database service |
+| Redis | 26379 | 6379 | Cache service |
+| Weaviate | 28080 | 28080 | Vector database |
+| MinIO API | 29000 | 9000 | Object storage API |
+| MinIO Console | 29090 | 9090 | Object storage console |

-### 镜像仓库
+### Image Registry

-所有镜像托管在阿里云容器镜像服务：
+Application images are published to GitHub Container Registry (GHCR) by GitHub Actions when a release tag such as `v3.1.0` is pushed:

 ```
-crpi-31mraxd99y2gqdgr.cn-beijing.personal.cr.aliyuncs.com/ruoyi_ai
+ghcr.io/ageerle/ruoyi-ai-backend:v3.1.0
+ghcr.io/ageerle/ruoyi-ai-mysql:v3.1.0
+ghcr.io/ageerle/ruoyi-ai-admin:v3.1.0
+ghcr.io/ageerle/ruoyi-ai-web:v3.1.0
 ```

-可用镜像：
- `mysql:v3` - MySQL 数据库（包含初始化 SQL）
- `redis:6.2` - Redis 缓存
- `weaviate:1.30.0` - 向量数据库
- `minio:latest` - 对象存储
- `ruoyi-ai-backend:latest` - 后端服务
- `ruoyi-ai-admin:latest` - 管理端前端
- `ruoyi-ai-web:latest` - 用户端前端
+The Compose file defaults to `latest`. To pin a release, create `docs/docker/ruoyi-ai/.env` with `RUIYI_VERSION=v3.1.0`.

-### 常用命令
+After the first workflow run, set the four GHCR packages to `Public` in GitHub so deployment servers can pull them without logging in.
+
+The same release tag must exist in `ageerle/ruoyi-admin` and `ageerle/ruoyi-web`; the publish workflow checks out those repositories at the backend release tag before building their images.
+
+### Common Commands

 ```bash
-# 停止所有服务
-docker-compose -f docker-compose-all.yaml down
+# Stop all services
+docker compose -f docs/docker/ruoyi-ai/docker-compose-all.yaml down

-# 查看服务日志
-docker-compose -f docker-compose-all.yaml logs -f [服务名]
+# View service logs
+docker compose -f docs/docker/ruoyi-ai/docker-compose-all.yaml logs -f [service-name]

-# 重启某个服务
-docker-compose -f docker-compose-all.yaml restart [服务名]
+# Restart a service
+docker compose -f docs/docker/ruoyi-ai/docker-compose-all.yaml restart [service-name]
 ```

-## 📚 使用文档
+## 📚 Documentation

-想要深入了解安装部署、功能配置和二次开发？
+Want to learn more about installation, deployment, configuration, and secondary development?

-**👉 [完整使用文档](https://doc.ruoyiai.chat/)**
+**👉 [Complete Documentation](https://doc.ruoyiai.chat/)**

-## 🤝 参与贡献
+## 🤝 Contributing

-我们热烈欢迎社区贡献！无论您是资深开发者还是初学者，都可以为项目贡献力量 💪
+We warmly welcome community contributions! Whether you are a seasoned developer or just getting started, you can contribute to the project 💪

-### 贡献方式
+### How to Contribute

-1. **Fork** 项目到您的账户
-2. **创建分支** (`git checkout -b feature/新功能名称`)
-3. **提交代码** (`git commit -m '添加某某功能'`)
-4. **推送分支** (`git push origin feature/新功能名称`)
-5. **发起 Pull Request**
+1. **Fork** the project to your account
+2. **Create a branch** (`git checkout -b feature/new-feature-name`)
+3. **Commit your changes** (`git commit -m 'Add new feature'`)
+4. **Push to the branch** (`git push origin feature/new-feature-name`)
+5. **Create a Pull Request**

-> 💡 **小贴士**：建议将 PR 提交到 GitHub，我们会自动同步到其他代码托管平台
+> 💡 **Tip**: We recommend submitting PRs to GitHub, we will automatically sync to other code hosting platforms

-## 📄 开源协议
+## 📄 License

-本项目采用 **MIT 开源协议**，详情请查看 [LICENSE](LICENSE) 文件。
+This project is licensed under the **MIT License**. See the [LICENSE](LICENSE) file for details.

-## 🙏 特别鸣谢
+## 🙏 Acknowledgments

-感谢以下优秀的开源项目为本项目提供支持：
- [Langchain4j](https://github.com/langchain4j/langchain4j) - 强大的 Java LLM 开发框架
- [RuoYi-Vue-Plus](https://gitee.com/dromara/RuoYi-Vue-Plus) - 成熟的企业级快速开发框架
- [Vben Admin](https://github.com/vbenjs/vue-vben-admin) - 现代化的 Vue 后台管理模板
+Thanks to the following excellent open-source projects for their support:
+- [Langchain4j](https://github.com/langchain4j/langchain4j) - Powerful Java LLM development framework
+- [RuoYi-Vue-Plus](https://gitee.com/dromara/RuoYi-Vue-Plus) - Mature enterprise-level rapid development framework
+- [Vben Admin](https://github.com/vbenjs/vue-vben-admin) - Modern Vue admin template

+## 💎 Sponsors

-## 💎 赞助商
-
-**感谢以下赞助商对本项目的支持：**
+**Thanks to the following sponsors for supporting this project:**

 <a href="https://www.atlascloud.ai?ref=89F97E">
  <img src="docs/image/sponsor/atlascloud_banner.png" alt="Atlas Cloud" width="160" height="80">
 </a>

-[访问Atlas Cloud官网](https://www.atlascloud.ai?ref=89F97E&utm_source=github&utm_campaign=ruoyi-drama) · [编程计划优惠](https://www.atlascloud.ai/console/coding-plan)
-全模态 AI 推理平台，为开发者提供统一的 AI API，支持视频生成、图像生成和大语言模型。一次接入，即可访问 **300+ 精选模型**。
+[Visit Atlas Cloud](https://www.atlascloud.ai?ref=89F97E&utm_source=github&utm_campaign=ruoyi-drama) · [Coding Plan Promotion](https://www.atlascloud.ai/console/coding-plan)
+A full-modal AI inference platform that gives developers a unified AI API, supporting video generation, image generation, and LLMs. Connect once to access **300+ curated models**.

 <a href="https://www.volcengine.com/activity/codingplan?utm_campaign=hw&utm_content=hw&utm_medium=devrel_tool_web&utm_source=OWO&utm_term=ageerle-ruoyi-ai">
-  <img src="docs/image/sponsor/huoshan.png" alt="火山引擎 CodingPlan" width="160" height="80">
+  <img src="docs/image/sponsor/huoshan.png" alt="Volcengine CodingPlan" width="160" height="80">
 </a>

-[注册即领2500万Tokens，立即前往](https://www.volcengine.com/activity/ai618?utm_campaign=hw&utm_content=hw&utm_medium=devrel_tool_web&utm_source=OWO&utm_term=ageerle-ruoyi-ai)
-享字节自研豆包模型+满血版开源 SOTA模型，覆盖文本、VLM、图像生成，全模态一站配齐：Seed-2.1、Seedream-5.0、GLM-5.2、DeepSeek等。不止编程、更能解决 Agent 复杂长程任务！
+[Sign up to claim 25 million tokens — go now](https://www.volcengine.com/activity/ai618?utm_campaign=hw&utm_content=hw&utm_medium=devrel_tool_web&utm_source=OWO&utm_term=ageerle-ruoyi-ai)
+Enjoy ByteDance's in-house Doubao models plus full-power open-source SOTA models, covering text, VLM, and image generation — all modalities in one stop: Seed-2.1, Seedream-5.0, GLM-5.2, DeepSeek, and more. Not just for coding — it can also tackle complex long-horizon Agent tasks!

-
-## 💬 社区交流
+## 💬 Community Chat

 <div align="center">

 <table>
 <tr>
 <td align="center">
-<img src="docs/image/wx.png" alt="微信二维码" width="200" height="200"><br>
-<strong>扫码添加作者微信</strong><br>
-<em>邀请进群学习</em>
+<img src="docs/image/wx.png" alt="WeChat QR Code" width="200" height="200"><br>
+<strong>Scan to add author on WeChat</strong><br>
+<em>Join group for learning</em>
 </td>
 <td align="center">
-<img src="docs/image/wx06.png" alt="微信二维码" width="200" height="200"><br>
-<strong>微信技术交流群</strong><br>
-<em>技术讨论</em>
+<img src="docs/image/wx06.png" alt="WeChat QR Code" width="200" height="200"><br>
+<strong>WeChat Tech Exchange Group</strong><br>
+<em>Technical discussion</em>
 </td>
 <td align="center">
-<img src="docs/image/qq.png" alt="QQ群二维码" width="200" height="200"><br>
-<strong>QQ技术交流群</strong><br>
-<em>技术讨论</em>
+<img src="docs/image/qq.png" alt="QQ Group QR Code" width="200" height="200"><br>
+<strong>QQ Tech Exchange Group</strong><br>
+<em>Technical discussion</em>
 </td>

 </tr>
@@ -255,11 +250,12 @@ docker-compose -f docker-compose-all.yaml restart [服务名]
 </div>

 ---
+
 <div align="center">

-**[⭐ 点个Star支持一下](https://github.com/ageerle/ruoyi-ai)** • **[ Fork 开始贡献](https://github.com/ageerle/ruoyi-ai/fork)** • **[📚 English](README_EN.md)** • **[📖 查看完整文档](https://doc.ruoyiai.chat/)**
+**[⭐ Star to Support](https://github.com/ageerle/ruoyi-ai)** • **[Fork to Contribute](https://github.com/ageerle/ruoyi-ai/fork)** • **[📚 中文](README_ZH.md)** • **[📖 Complete Documentation](https://doc.ruoyiai.chat/)**

-*用 ❤️ 打造，由 RuoYi AI 开源社区维护*
+*Built with ❤️, maintained by the RuoYi AI open-source community*

 </div>

--- a/README_EN.md
+++ b/README_EN.md
@@ -1,310 +0,0 @@
-
-# RuoYi AI
-
-<div align="center">
-
-[![Contributors][contributors-shield]][contributors-url]
-[![Forks][forks-shield]][forks-url]
-[![Stargazers][stars-shield]][stars-url]
-[![Issues][issues-shield]][issues-url]
-[![MIT License][license-shield]][license-url]
-
-
-<p align="center">
-  <a href="https://trendshift.io/repositories/13209">
-    <img src="https://trendshift.io/api/badge/repositories/13209" alt="GitHub Trending">
-  </a>
-</p>
-
-<img src="docs/image/logo.png" alt="RuoYi AI Logo" width="120" height="120">
-
-### Enterprise-Grade AI Assistant Platform
-
-*An out-of-the-box full-stack AI platform supporting multi-agent collaboration, Supervisor mode orchestration, and multiple decision models, with advanced RAG technology and visual workflow orchestration capabilities*
-
-**[中文](README.md)** | **[📖 Documentation](https://doc.ruoyiai.chat/)** |
-**[🚀 Live Demo](https://web.ruoyiai.chat/)** | **[🐛 Report Issues](https://github.com/ageerle/ruoyi-ai/issues)** | **[💡 Feature Requests](https://github.com/ageerle/ruoyi-ai/issues)**
-
-</div>
-
-
-
-
-## ✨ Core Features
-
-| Module | Current Capabilities |
-|:---:|---|
-| **Model Management** | Multi-model integration (OpenAI/DeepSeek/Tongyi/Zhipu/MiniMax), multi-modal understanding, Coze/DIFY/FastGPT platform integration |
-| **Knowledge Base** | Local RAG + Vector DB (Milvus/Weaviate/Qdrant) + Document parsing |
-| **Tool Management** | MCP protocol integration, Skills capability + Extensible tool ecosystem |
-| **Workflow Orchestration** | Visual workflow designer, drag-and-drop node orchestration, SSE streaming execution, currently supports model calls, email sending, manual review nodes |
-| **Multi-Agent** | Agent framework based on Langchain4j, Supervisor mode orchestration, supports multiple decision models |
-
-### Project Repositories
-
-| Module     | GitHub Repository                                             | Gitee Repository                                             | GitCode Repository                                             |
-|----------|-------------------------------------------------------|------------------------------------------------------|--------------------------------------------------------|
-| 🔧 Backend  | [ruoyi-ai](https://github.com/ageerle/ruoyi-ai)       | [ruoyi-ai](https://gitee.com/ageerle/ruoyi-ai)       | [ruoyi-ai](https://gitcode.com/ageerle/ruoyi-ai)       |
-| 🎨 User Frontend  | [ruoyi-web](https://github.com/ageerle/ruoyi-web)     | [ruoyi-web](https://gitee.com/ageerle/ruoyi-web)     | [ruoyi-web](https://gitcode.com/ageerle/ruoyi-web)     |
-| 🛠️ Admin Panel | [ruoyi-admin](https://github.com/ageerle/ruoyi-admin) | [ruoyi-admin](https://gitee.com/ageerle/ruoyi-admin) | [ruoyi-admin](https://gitcode.com/ageerle/ruoyi-admin) |
-| 🎬 Drama | [ruoyi-drama](https://github.com/ageerle/ruoyi-drama) | [ruoyi-drama](https://gitee.com/ageerle/ruoyi-drama) | |
-| 🤖 Copilot | [ruoyi-copilot](https://github.com/ageerle/ruoyi-copilot) | [ruoyi-copilot](https://gitee.com/ageerle/ruoyi-copilot) | [ruoyi-copilot](https://gitcode.com/ageerle/ruoyi-copilot) |
-| 📱 Mini-App | [ruoyi-uniapp](https://github.com/ageerle/ruoyi-uniapp) | [ruoyi-uniapp](https://gitee.com/ageerle/ruoyi-uniapp) | [ruoyi-uniapp](https://gitcode.com/ageerle/ruoyi-uniapp) |
-
-### Partner Projects
-| Project Name           | GitHub Repository                                             | Gitee Repository |
-|----------------|-------------------------------------------------------|------------------------------------------------------|
-| element-plus-x | [element-plus-x](https://github.com/element-plus-x/Element-Plus-X)       | [element-plus-x](https://gitee.com/he-jiayue/element-plus-x)       |
-
-## 🛠️ Technical Architecture
-
-### Core Framework
- **Backend**: Spring Boot 3.5.8 + Langchain4j
- **Data Storage**: MySQL 8.0 + Redis + Vector Databases (Milvus/Weaviate/Qdrant)
- **Frontend**: Vue 3 + Vben Admin + element-plus-x
- **Security**: Sa-Token + JWT dual-layer security
- **Document Processing**: PDF, Word, Excel parsing, intelligent image analysis
- **Real-time Communication**: WebSocket real-time communication, SSE streaming response
- **System Monitoring**: Comprehensive logging system, performance monitoring, service health checks
-
-## 🐳 Docker Deployment
-
-This project provides two Docker deployment methods:
-
-### Method 1: One-click Start All Services (Recommended)
-
-Use `docker-compose-all.yaml` to start all services at once (including backend, admin panel, user frontend, and dependencies):
-
-```bash
-# Clone the repository
-git clone https://github.com/ageerle/ruoyi-ai.git
-cd ruoyi-ai
-
-# Start all services (pull pre-built images from registry)
-docker-compose -f docker-compose-all.yaml up -d
-
-# Check service status
-docker-compose -f docker-compose-all.yaml ps
-
-# Access services
-# Admin Panel: http://localhost:25666 (admin / admin123)
-# User Frontend: http://localhost:25137
-# Backend API: http://localhost:26039
-```
-
-### Method 2: Step-by-step Deployment (Source Build)
-
-If you need to build backend services from source, follow these steps:
-
-#### Step 1: Deploy Backend Service
-
-```bash
-# Enter backend project directory
-cd ruoyi-ai
-
-# Start backend service (build from source)
-docker-compose up -d --build
-
-# Wait for backend service to start
-docker-compose logs -f backend
-```
-
-#### Step 2: Deploy Admin Panel
-
-```bash
-# Enter admin panel project directory
-cd ruoyi-admin
-
-# Build and start admin panel
-docker-compose up -d --build
-
-# Access admin panel
-# URL: http://localhost:5666
-```
-
-#### Step 3: Deploy User Frontend (Optional)
-
-```bash
-# Enter user frontend project directory
-cd ruoyi-web
-
-# Build and start user frontend
-docker-compose up -d --build
-
-# Access user frontend
-# URL: http://localhost:5137
-```
-
-### Service Ports
-
-| Service | One-click Port | Step-by-step Port | Description |
-|------|-------------|-------------|------|
-| Admin Panel | 25666 | 5666 | Admin backend access |
-| User Frontend | 25137 | 5137 | User frontend access |
-| Backend Service | 26039 | 6039 | Backend API service |
-| MySQL | 23306 | 23306 | Database service |
-| Redis | 26379 | 6379 | Cache service |
-| Weaviate | 28080 | 28080 | Vector database |
-| MinIO API | 29000 | 9000 | Object storage API |
-| MinIO Console | 29090 | 9090 | Object storage console |
-
-### Image Registry
-
-All images are hosted on Alibaba Cloud Container Registry:
-
-```
-crpi-31mraxd99y2gqdgr.cn-beijing.personal.cr.aliyuncs.com/ruoyi_ai
-```
-
-Available images:
- `mysql:v3` - MySQL database (includes initialization SQL)
- `redis:6.2` - Redis cache
- `weaviate:1.30.0` - Vector database
- `minio:latest` - Object storage
- `ruoyi-ai-backend:latest` - Backend service
- `ruoyi-ai-admin:latest` - Admin frontend
- `ruoyi-ai-web:latest` - User frontend
-
-### Common Commands
-
-```bash
-# Stop all services
-docker-compose -f docker-compose-all.yaml down
-
-# View service logs
-docker-compose -f docker-compose-all.yaml logs -f [service-name]
-
-# Restart a service
-docker-compose -f docker-compose-all.yaml restart [service-name]
-```
-
-### MiniMax Configuration
-
-The built-in MiniMax provider accepts one API Host value and selects the matching protocol adapter. Use a Base URL from this table:
-
-| Region | OpenAI-compatible Base URL | Anthropic-compatible Base URL |
-| --- | --- | --- |
-| Global | `https://api.minimax.io/v1` | `https://api.minimax.io/anthropic` |
-| China | `https://api.minimaxi.com/v1` | `https://api.minimaxi.com/anthropic` |
-
-For Anthropic-compatible requests, configure the Base URL ending in `/anthropic`. Do not append `/v1` or `/v1/messages`; the provider adapter derives the request path internally.
-
-| Model ID | Total context | Input modalities | Thinking |
-| --- | ---: | --- | --- |
-| `MiniMax-M3` | 1,000,000 tokens | Text, image, video | Adaptive or disabled |
-| `MiniMax-M2.7` | 204,800 tokens | Text | Always on |
-
-Current pay-as-you-go prices are in USD per million tokens:
-
-| Model | Service tier and input range | Input | Output | Cache read | Cache write |
-| --- | --- | ---: | ---: | ---: | ---: |
-| `MiniMax-M3` | Standard, up to 512,000 input tokens | $0.30 | $1.20 | $0.06 | Not listed |
-| `MiniMax-M3` | Standard, over 512,000 input tokens | $0.60 | $2.40 | $0.12 | Not listed |
-| `MiniMax-M3` | Priority, up to 512,000 input tokens | $0.45 | $1.80 | $0.09 | Not listed |
-| `MiniMax-M3` | Priority, over 512,000 input tokens | $0.90 | $3.60 | $0.18 | Not listed |
-| `MiniMax-M2.7` | Standard | $0.30 | $1.20 | $0.06 | $0.375 |
-
-See the [official API overview](https://platform.minimax.io/docs/api-reference/api-overview) and [pay-as-you-go pricing](https://platform.minimax.io/docs/guides/pricing-paygo) for current details.
-
-## 📚 Documentation
-
-Want to learn more about installation, deployment, configuration, and secondary development?
-
-**👉 [Complete Documentation](https://doc.ruoyiai.chat/)**
-
-## 🤝 Contributing
-
-We warmly welcome community contributions! Whether you are a seasoned developer or just getting started, you can contribute to the project 💪
-
-### How to Contribute
-
-1. **Fork** the project to your account
-2. **Create a branch** (`git checkout -b feature/new-feature-name`)
-3. **Commit your changes** (`git commit -m 'Add new feature'`)
-4. **Push to the branch** (`git push origin feature/new-feature-name`)
-5. **Create a Pull Request**
-
-> 💡 **Tip**: We recommend submitting PRs to GitHub, we will automatically sync to other code hosting platforms
-
-## 📄 License
-
-This project is licensed under the **MIT License**. See the [LICENSE](LICENSE) file for details.
-
-## 🙏 Acknowledgments
-
-Thanks to the following excellent open-source projects for their support:
- [Langchain4j](https://github.com/langchain4j/langchain4j) - Powerful Java LLM development framework
- [RuoYi-Vue-Plus](https://gitee.com/dromara/RuoYi-Vue-Plus) - Mature enterprise-level rapid development framework
- [Vben Admin](https://github.com/vbenjs/vue-vben-admin) - Modern Vue admin template
-
-## 💎 Sponsors
-
-**Thanks to the following sponsors for supporting this project:**
-
-<a href="https://www.atlascloud.ai?ref=89F97E">
-  <img src="docs/image/sponsor/atlascloud_banner.png" alt="Atlas Cloud" width="160" height="80">
-</a>
-
-[Visit Atlas Cloud](https://www.atlascloud.ai?ref=89F97E) · [Coding Plan Promotion](https://www.atlascloud.ai/console/coding-plan)
-A full-modal AI inference platform that gives developers a unified AI API, supporting video generation, image generation, and LLMs. Connect once to access **300+ curated models**.
-
-<a href="https://www.volcengine.com/activity/codingplan?utm_campaign=hw&utm_content=hw&utm_medium=devrel_tool_web&utm_source=OWO&utm_term=ageerle-ruoyi-ai">
-  <img src="docs/image/sponsor/huoshan.png" alt="Volcengine CodingPlan" width="160" height="80">
-</a>
-
-[Volcengine CodingPlan Developer Program](https://www.volcengine.com/activity/codingplan?utm_campaign=hw&utm_content=hw&utm_medium=devrel_tool_web&utm_source=OWO&utm_term=ageerle-ruoyi-ai)
-Volcengine is ByteDance's cloud and AI service platform. Volcengine Ark provides API access to Doubao LLM, DeepSeek, and more — a one-stop AI development and inference platform for developers.
-
-## 💬 Community Chat
-
-<div align="center">
-
-<table>
-<tr>
-<td align="center">
-<img src="docs/image/wx.png" alt="WeChat QR Code" width="200" height="200"><br>
-<strong>Scan to add author on WeChat</strong><br>
-<em>Join group for learning</em>
-</td>
-<td align="center">
-<img src="docs/image/qq.png" alt="QQ Group QR Code" width="200" height="200"><br>
-<strong>QQ Tech Exchange Group</strong><br>
-<em>Technical discussion</em>
-</td>
-
-</tr>
-</table>
-
-</div>
-
---
-
-<div align="center">
-
-**[⭐ Star to Support](https://github.com/ageerle/ruoyi-ai)** • **[Fork to Contribute](https://github.com/ageerle/ruoyi-ai/fork)** • **[📚 中文](README.md)** • **[📖 Complete Documentation](https://doc.ruoyiai.chat/)**
-
-*Built with ❤️, maintained by the RuoYi AI open-source community*
-
-</div>
-
-<!-- Badge Links -->
-
-[contributors-shield]: https://img.shields.io/github/contributors/ageerle/ruoyi-ai.svg?style=flat-square
-
-[contributors-url]: https://github.com/ageerle/ruoyi-ai/graphs/contributors
-
-[forks-shield]: https://img.shields.io/github/forks/ageerle/ruoyi-ai.svg?style=flat-square
-
-[forks-url]: https://github.com/ageerle/ruoyi-ai/network/members
-
-[stars-shield]: https://img.shields.io/github/stars/ageerle/ruoyi-ai.svg?style=flat-square
-
-[stars-url]: https://github.com/ageerle/ruoyi-ai/stargazers
-
-[issues-shield]: https://img.shields.io/github/issues/ageerle/ruoyi-ai.svg?style=flat-square
-
-[issues-url]: https://github.com/ageerle/ruoyi-ai/issues
-
-[license-shield]: https://img.shields.io/github/license/ageerle/ruoyi-ai.svg?style=flat-square
-
-[license-url]: https://github.com/ageerle/ruoyi-ai/blob/main/LICENSE
--- a/README_ZH.md
+++ b/README_ZH.md
@@ -0,0 +1,282 @@
+# RuoYi AI
+
+<div align="center">
+
+[![Contributors][contributors-shield]][contributors-url] [![Forks][forks-shield]][forks-url] [![Stargazers][stars-shield]][stars-url] [![Issues][issues-shield]][issues-url] [![MIT License][license-shield]][license-url]
+
+
+<p align="center">
+  <a href="https://trendshift.io/repositories/13209">
+    <img src="https://trendshift.io/api/badge/repositories/13209" alt="GitHub Trending">
+  </a>
+</p>
+
+<img src="docs/image/logo.png" alt="RuoYi AI Logo" width="120" height="120">
+
+### 企业级AI助手平台
+
+*开箱即用的全栈AI平台，支持多智能体协同、Supervisor模式编排、多种决策模式、RAG技术和流程编排能力*
+
+**[English](README.md)** | **[📖 使用文档](https://doc.ruoyiai.chat/)** |
+**[🚀 在线体验](https://web.ruoyiai.chat/)** | **[🐛 问题反馈](https://github.com/ageerle/ruoyi-ai/issues)** | **[💡 功能建议](https://github.com/ageerle/ruoyi-ai/issues)**
+
+</div>
+
+
+## ✨ 核心亮点
+
+|    模块     | 现有能力
+|:---------:|---
+| **模型管理**  | 多模型接入(DeepSeek/智谱/MIMO/百炼/OpenAI)、多模态理解、Coze/DIFY/FastGPT/RAGFlow平台集成
+| **知识管理**  | 本地RAG + 向量库(Milvus/Weaviate/Qdrant)  + 文档解析
+| **工具管理**  | Mcp协议集成、Skills能力 + 可扩展工具生态
+| **流程编排**  | 可视化工作流设计器、节点拖拽编排、SSE流式执行,目前已经支持模型调用,邮件发送,人工审核等节点
+| **智能体管理** | 基于Langchain4j的Agent框架、Supervisor模式编排,支持多种决策模型,可以灵活搭配工具,skills
+
+
+### 项目源码
+
+| 项目模块     | GitHub 仓库                                             | Gitee 仓库                                             | GitCode 仓库                                             |
+|----------|-------------------------------------------------------|------------------------------------------------------|--------------------------------------------------------|
+| 🔧 后端服务  | [ruoyi-ai](https://github.com/ageerle/ruoyi-ai)       | [ruoyi-ai](https://gitee.com/ageerle/ruoyi-ai)       | [ruoyi-ai](https://gitcode.com/ageerle/ruoyi-ai)       |
+| 🎨 用户前端  | [ruoyi-web](https://github.com/ageerle/ruoyi-web)     | [ruoyi-web](https://gitee.com/ageerle/ruoyi-web)     | [ruoyi-web](https://gitcode.com/ageerle/ruoyi-web)     |
+| 🛠️ 管理后台 | [ruoyi-admin](https://github.com/ageerle/ruoyi-admin) | [ruoyi-admin](https://gitee.com/ageerle/ruoyi-admin) | [ruoyi-admin](https://gitcode.com/ageerle/ruoyi-admin) |
+| 🎬 短剧平台  | [ruoyi-drama](https://github.com/ageerle/ruoyi-drama) | [ruoyi-drama](https://gitee.com/ageerle/ruoyi-drama) | [ruoyi-drama](https://gitcode.com/ageerle/ruoyi-drama)   |
+| 🤖 编程助手  | [ruoyi-copilot](https://github.com/ageerle/ruoyi-copilot) | [ruoyi-copilot](https://gitee.com/ageerle/ruoyi-copilot) | [ruoyi-copilot](https://gitcode.com/ageerle/ruoyi-copilot) |
+| 📱 小程序端  | [ruoyi-uniapp](https://github.com/ageerle/ruoyi-uniapp) | [ruoyi-uniapp](https://gitee.com/ageerle/ruoyi-uniapp) | [ruoyi-uniapp](https://gitcode.com/ageerle/ruoyi-uniapp) |
+
+### 合作项目
+| 项目名称           | GitHub 仓库                                             | Gitee 仓库
+|----------------|-------------------------------------------------------|------------------------------------------------------|
+| element-plus-x | [element-plus-x](https://github.com/element-plus-x/Element-Plus-X)       | [element-plus-x](https://gitee.com/he-jiayue/element-plus-x)       |
+
+## 🛠️ 技术架构
+
+### 核心框架
+- **后端架构**：Spring Boot 3.5.8 + Langchain4j
+- **数据存储**：MySQL 8.0 + Redis + 向量数据库（Milvus/Weaviate/Qdrant）
+- **前端技术**：Vue 3 + Vben Admin + element-plus-x
+- **安全认证**：Sa-Token + JWT 双重保障
+- **文档处理**：PDF、Word、Excel 解析，图像智能分析
+- **实时通信**：WebSocket 实时通信，SSE 流式响应
+- **系统监控**：完善的日志体系、性能监控、服务健康检查
+
+## 🐳 Docker 部署
+
+本项目提供两种 Docker 部署方式：
+
+### 方式一：一键启动所有服务（推荐）
+
+使用 `docker-compose-all.yaml` 可以一键启动所有服务（包括后端、管理端、用户端及依赖服务）：
+
+```bash
+# 克隆仓库
+git clone https://github.com/ageerle/ruoyi-ai.git
+cd ruoyi-ai
+
+# 启动所有服务（从 GHCR 拉取预构建镜像）
+docker compose -f docs/docker/ruoyi-ai/docker-compose-all.yaml up -d
+
+# 查看服务状态
+docker compose -f docs/docker/ruoyi-ai/docker-compose-all.yaml ps
+
+# 访问服务
+# 管理端: http://localhost:25666 (admin / admin123)
+# 用户端: http://localhost:25137
+# 后端API: http://localhost:26039
+```
+
+### 方式二：分步部署（源码编译）
+
+如果您需要从源码构建后端服务，请按照以下步骤操作：
+
+#### 第一步：部署后端服务
+
+```bash
+# 进入后端项目目录
+cd ruoyi-ai
+
+# 启动后端服务（源码编译构建）
+docker-compose up -d --build
+
+# 等待后端服务启动完成
+docker-compose logs -f backend
+```
+
+#### 第二步：部署管理端
+
+```bash
+# 进入管理端项目目录
+cd ruoyi-admin
+
+# 构建并启动管理端
+docker-compose up -d --build
+
+# 访问管理端
+# 地址: http://localhost:5666
+```
+
+#### 第三步：部署用户端（可选）
+
+```bash
+# 进入用户端项目目录
+cd ruoyi-web
+
+# 构建并启动用户端
+docker-compose up -d --build
+
+# 访问用户端
+# 地址: http://localhost:5137
+```
+
+### 服务端口说明
+
+| 服务 | 一键启动端口 | 分步部署端口 | 说明 |
+|------|-------------|-------------|------|
+| 管理端 | 25666 | 5666 | 管理后台访问地址 |
+| 用户端 | 25137 | 5137 | 用户前端访问地址 |
+| 后端服务 | 26039 | 6039 | 后端 API 服务 |
+| MySQL | 23306 | 23306 | 数据库服务 |
+| Redis | 26379 | 6379 | 缓存服务 |
+| Weaviate | 28080 | 28080 | 向量数据库 |
+| MinIO API | 29000 | 9000 | 对象存储 API |
+| MinIO Console | 29090 | 9090 | 对象存储控制台 |
+
+### 镜像仓库
+
+每次推送类似 `v3.1.0` 的版本标签后，GitHub Actions 会自动构建并发布应用镜像到 GitHub Container Registry（GHCR）：
+
+```
+ghcr.io/ageerle/ruoyi-ai-backend:v3.1.0
+ghcr.io/ageerle/ruoyi-ai-mysql:v3.1.0
+ghcr.io/ageerle/ruoyi-ai-admin:v3.1.0
+ghcr.io/ageerle/ruoyi-ai-web:v3.1.0
+```
+
+Compose 默认使用 `latest`。如需固定版本，可在 `docs/docker/ruoyi-ai/.env` 中设置 `RUIYI_VERSION=v3.1.0`。
+
+首次工作流运行完成后，请在 GitHub 的 Packages 设置中将这四个 GHCR 镜像设为 `Public`，用户服务器才能免登录拉取。
+
+管理端 `ageerle/ruoyi-admin` 和用户端 `ageerle/ruoyi-web` 也必须存在同名版本标签；发布工作流会使用后端发布标签检出这两个仓库后再构建镜像。
+
+### 常用命令
+
+```bash
+# 停止所有服务
+docker compose -f docs/docker/ruoyi-ai/docker-compose-all.yaml down
+
+# 查看服务日志
+docker compose -f docs/docker/ruoyi-ai/docker-compose-all.yaml logs -f [服务名]
+
+# 重启某个服务
+docker compose -f docs/docker/ruoyi-ai/docker-compose-all.yaml restart [服务名]
+```
+
+## 📚 使用文档
+
+想要深入了解安装部署、功能配置和二次开发？
+
+**👉 [完整使用文档](https://doc.ruoyiai.chat/)**
+
+## 🤝 参与贡献
+
+我们热烈欢迎社区贡献！无论您是资深开发者还是初学者，都可以为项目贡献力量 💪
+
+### 贡献方式
+
+1. **Fork** 项目到您的账户
+2. **创建分支** (`git checkout -b feature/新功能名称`)
+3. **提交代码** (`git commit -m '添加某某功能'`)
+4. **推送分支** (`git push origin feature/新功能名称`)
+5. **发起 Pull Request**
+
+> 💡 **小贴士**：建议将 PR 提交到 GitHub，我们会自动同步到其他代码托管平台
+
+## 📄 开源协议
+
+本项目采用 **MIT 开源协议**，详情请查看 [LICENSE](LICENSE) 文件。
+
+## 🙏 特别鸣谢
+
+感谢以下优秀的开源项目为本项目提供支持：
+- [Langchain4j](https://github.com/langchain4j/langchain4j) - 强大的 Java LLM 开发框架
+- [RuoYi-Vue-Plus](https://gitee.com/dromara/RuoYi-Vue-Plus) - 成熟的企业级快速开发框架
+- [Vben Admin](https://github.com/vbenjs/vue-vben-admin) - 现代化的 Vue 后台管理模板
+
+
+## 💎 赞助商
+
+**感谢以下赞助商对本项目的支持：**
+
+<a href="https://www.atlascloud.ai?ref=89F97E">
+  <img src="docs/image/sponsor/atlascloud_banner.png" alt="Atlas Cloud" width="160" height="80">
+</a>
+
+[访问Atlas Cloud官网](https://www.atlascloud.ai?ref=89F97E&utm_source=github&utm_campaign=ruoyi-drama) · [编程计划优惠](https://www.atlascloud.ai/console/coding-plan)
+全模态 AI 推理平台，为开发者提供统一的 AI API，支持视频生成、图像生成和大语言模型。一次接入，即可访问 **300+ 精选模型**。
+
+<a href="https://www.volcengine.com/activity/codingplan?utm_campaign=hw&utm_content=hw&utm_medium=devrel_tool_web&utm_source=OWO&utm_term=ageerle-ruoyi-ai">
+  <img src="docs/image/sponsor/huoshan.png" alt="火山引擎 CodingPlan" width="160" height="80">
+</a>
+
+[注册即领2500万Tokens，立即前往](https://www.volcengine.com/activity/ai618?utm_campaign=hw&utm_content=hw&utm_medium=devrel_tool_web&utm_source=OWO&utm_term=ageerle-ruoyi-ai)
+享字节自研豆包模型+满血版开源 SOTA模型，覆盖文本、VLM、图像生成，全模态一站配齐：Seed-2.1、Seedream-5.0、GLM-5.2、DeepSeek等。不止编程、更能解决 Agent 复杂长程任务！
+
+
+## 💬 社区交流
+
+<div align="center">
+
+<table>
+<tr>
+<td align="center">
+<img src="docs/image/wx.png" alt="微信二维码" width="200" height="200"><br>
+<strong>扫码添加作者微信</strong><br>
+<em>邀请进群学习</em>
+</td>
+<td align="center">
+<img src="docs/image/wx06.png" alt="微信二维码" width="200" height="200"><br>
+<strong>微信技术交流群</strong><br>
+<em>技术讨论</em>
+</td>
+<td align="center">
+<img src="docs/image/qq.png" alt="QQ群二维码" width="200" height="200"><br>
+<strong>QQ技术交流群</strong><br>
+<em>技术讨论</em>
+</td>
+
+</tr>
+</table>
+
+</div>
+
+---
+<div align="center">
+
+**[⭐ 点个Star支持一下](https://github.com/ageerle/ruoyi-ai)** • **[ Fork 开始贡献](https://github.com/ageerle/ruoyi-ai/fork)** • **[📚 English](README.md)** • **[📖 查看完整文档](https://doc.ruoyiai.chat/)**
+
+*用 ❤️ 打造，由 RuoYi AI 开源社区维护*
+
+</div>
+
+<!-- Badge Links -->
+
+[contributors-shield]: https://img.shields.io/github/contributors/ageerle/ruoyi-ai.svg?style=flat-square
+
+[contributors-url]: https://github.com/ageerle/ruoyi-ai/graphs/contributors
+
+[forks-shield]: https://img.shields.io/github/forks/ageerle/ruoyi-ai.svg?style=flat-square
+
+[forks-url]: https://github.com/ageerle/ruoyi-ai/network/members
+
+[stars-shield]: https://img.shields.io/github/stars/ageerle/ruoyi-ai.svg?style=flat-square
+
+[stars-url]: https://github.com/ageerle/ruoyi-ai/stargazers
+
+[issues-shield]: https://img.shields.io/github/issues/ageerle/ruoyi-ai.svg?style=flat-square
+
+[issues-url]: https://github.com/ageerle/ruoyi-ai/issues
+
+[license-shield]: https://img.shields.io/github/license/ageerle/ruoyi-ai.svg?style=flat-square
+
+[license-url]: https://github.com/ageerle/ruoyi-ai/blob/main/LICENSE
--- a/docs/docker/ruoyi-ai/.env.example
+++ b/docs/docker/ruoyi-ai/.env.example
@@ -0,0 +1,6 @@
+# GitHub Container Registry namespace owner.
+# For the upstream project this is ageerle; change it when using a fork.
+IMAGE_OWNER=ageerle
+
+# Use latest for convenience, or pin a release such as v3.1.0.
+RUIYI_VERSION=latest
--- a/docs/docker/ruoyi-ai/docker-compose-all.yaml
+++ b/docs/docker/ruoyi-ai/docker-compose-all.yaml
@@ -10,13 +10,13 @@
 #   - RuoYi-Admin (管理端前端)
 #   - RuoYi-Web (用户端前端)
 #
-# 镜像仓库地址: crpi-31mraxd99y2gqdgr.cn-beijing.personal.cr.aliyuncs.com/ruoyi_ai
+# 镜像仓库地址: ghcr.io/ageerle

 services:
  # ==================== MySQL 数据库 ====================
  mysql:
-    # 阿里云镜像地址（包含初始化SQL）
-    image: crpi-31mraxd99y2gqdgr.cn-beijing.personal.cr.aliyuncs.com/ruoyi_ai/mysql:v3
+    # GHCR 镜像（包含初始化SQL）
+    image: ghcr.io/${IMAGE_OWNER:-ageerle}/ruoyi-ai-mysql:${RUIYI_VERSION:-latest}
    container_name: ruoyi-ai-mysql
    restart: always
    ports:
@@ -41,7 +41,7 @@ services:

  # ==================== Redis 缓存 ====================
  redis:
-    image: crpi-31mraxd99y2gqdgr.cn-beijing.personal.cr.aliyuncs.com/ruoyi_ai/redis:6.2
+    image: redis:6.2
    container_name: ruoyi-ai-redis
    restart: always
    ports:
@@ -59,7 +59,7 @@ services:

  # ==================== Weaviate 向量数据库 ====================
  weaviate:
-    image: crpi-31mraxd99y2gqdgr.cn-beijing.personal.cr.aliyuncs.com/ruoyi_ai/weaviate:1.30.0
+    image: semitechnologies/weaviate:1.30.0
    container_name: ruoyi-ai-weaviate
    restart: always
    ports:
@@ -78,7 +78,7 @@ services:

  # ==================== MinIO 对象存储 ====================
  minio:
-    image: crpi-31mraxd99y2gqdgr.cn-beijing.personal.cr.aliyuncs.com/ruoyi_ai/minio:latest
+    image: minio/minio:latest
    container_name: ruoyi-ai-minio
    restart: always
    ports:
@@ -95,7 +95,7 @@ services:

  # ==================== RuoYi-AI 后端服务 ====================
  backend:
-    image: crpi-31mraxd99y2gqdgr.cn-beijing.personal.cr.aliyuncs.com/ruoyi_ai/ruoyi-ai-backend:latest
+    image: ghcr.io/${IMAGE_OWNER:-ageerle}/ruoyi-ai-backend:${RUIYI_VERSION:-latest}
    container_name: ruoyi-ai-backend
    restart: always
    ports:
@@ -129,7 +129,7 @@ services:

  # ==================== RuoYi-AI 管理端前端 ====================
  admin-frontend:
-    image: crpi-31mraxd99y2gqdgr.cn-beijing.personal.cr.aliyuncs.com/ruoyi_ai/ruoyi-ai-admin:latest
+    image: ghcr.io/${IMAGE_OWNER:-ageerle}/ruoyi-ai-admin:${RUIYI_VERSION:-latest}
    container_name: ruoyi-ai-admin
    restart: always
    ports:
@@ -154,7 +154,7 @@ services:

  # ==================== RuoYi-AI 用户端前端 ====================
  web-frontend:
-    image: crpi-31mraxd99y2gqdgr.cn-beijing.personal.cr.aliyuncs.com/ruoyi_ai/ruoyi-ai-web:latest
+    image: ghcr.io/${IMAGE_OWNER:-ageerle}/ruoyi-ai-web:${RUIYI_VERSION:-latest}
    container_name: ruoyi-ai-web
    restart: always
    ports:
--- a/docs/script/sql/ruoyi-ai.sql
+++ b/docs/script/sql/ruoyi-ai.sql
@@ -2291,11 +2291,10 @@ INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`,
 INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`) VALUES (2027193296990957569, '000000', '文生图节点响应模板', 'node.image.template', '🎨 文生图节点：结束响应 - 图片URL: ', 'Y', 103, 1, '2026-02-27 09:25:20', 1, '2026-02-27 09:31:52', NULL);
 INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`) VALUES (2027193820393959425, '000000', '发送邮箱节点响应模板', 'node.mailsend.template', '📧 发送邮箱节点：结束响应 - ', 'Y', 103, 1, '2026-02-27 09:27:25', 1, '2026-02-27 09:32:05', NULL);
 INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`) VALUES (2027194134438277122, '000000', '结束节点响应模板', 'node.end.template', '🔚 流程已执行完毕，如果您有其他需求，请随时重新发起请求。', 'Y', 103, 1, '2026-02-27 09:28:40', 1, '2026-02-27 09:32:53', NULL);
-INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`) VALUES (2027206492573335554, '000000', '人机交互节点响应模板', 'node.humanFeedback.template', '👤 人机交互节点：等待用户操作 - ', 'Y', 103, 1, '2026-02-27 10:17:46', 1, '2026-02-27 10:17:46', NULL);
 INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`) VALUES (2027208880369647617, '000000', '条件分支节点响应模板', 'node.switch.template', '🔀 条件分支节点：触发 -> 跳转到节点 ', 'Y', 103, 1, '2026-02-27 10:27:15', 1, '2026-02-27 10:35:54', NULL);
 INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`) VALUES (2027213914603995137, '000000', '大模型回答节点响应模板', 'node.llmAnswer.template', '🤖 LLM 节点 生成回答：', 'Y', 103, 1, '2026-02-27 10:47:16', 1, '2026-02-27 10:52:40', NULL);
-INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`) VALUES (2027214387000066050, '000000', '关键词提取响应模板', 'node.keywordExtractor.template', '🔑 关键词提取节点 处理完成 ： ', 'Y', 103, 1, '2026-02-27 10:49:08', 1, '2026-02-27 10:52:08', NULL);
 INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`) VALUES (2027217577397391361, '000000', '工作流异常响应模板', 'node.exception.template', '🛑 工作流发生异常：', 'N', 103, 1, '2026-02-27 11:01:49', 1, '2026-02-27 11:02:01', NULL);
+INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`) VALUES (2084157200000000003, '000000', '网络搜索节点响应模板', 'node.googleSearch.template', '🔍 网络搜索节点处理完成：', 'Y', 103, 1, '2026-07-29 19:40:00', 1, '2026-07-29 19:40:00', NULL);

 -- ----------------------------
 -- Table structure for sys_dept
@@ -3417,7 +3416,7 @@ CREATE TABLE `t_workflow_component`  (
                                         `tenant_id` varchar(20) CHARACTER SET utf8mb4 COLLATE utf8mb4_0900_ai_ci NULL DEFAULT '000000' COMMENT '租户编号',
                                         PRIMARY KEY (`id`) USING BTREE,
                                         INDEX `idx_display_order`(`display_order` ASC) USING BTREE
-) ENGINE = InnoDB AUTO_INCREMENT = 37 CHARACTER SET = utf8mb4 COLLATE = utf8mb4_0900_ai_ci COMMENT = '工作流组件库 | Workflow Component' ROW_FORMAT = DYNAMIC;
+) ENGINE = InnoDB AUTO_INCREMENT = 38 CHARACTER SET = utf8mb4 COLLATE = utf8mb4_0900_ai_ci COMMENT = '工作流组件库 | Workflow Component' ROW_FORMAT = DYNAMIC;

 -- ----------------------------
 -- Records of t_workflow_component
@@ -3425,9 +3424,8 @@ CREATE TABLE `t_workflow_component`  (
 INSERT INTO `t_workflow_component` VALUES (17, '5cd68dccbbb411f0bb7840c2ba9a7fbc', 'Start', '开始', '流程由此开始', 0, 1, '2025-11-07 16:32:49', '2025-11-07 16:32:49', 0, '000000');
 INSERT INTO `t_workflow_component` VALUES (18, '5cd6ac69bbb411f0bb7840c2ba9a7fbc', 'End', '结束', '流程由此结束', 0, 1, '2025-11-07 16:32:49', '2025-11-07 16:32:49', 0, '000000');
 INSERT INTO `t_workflow_component` VALUES (19, '5cd6c8eabbb411f0bb7840c2ba9a7fbc', 'Answer', '生成回答', '调用大语言模型回答问题', 0, 1, '2025-11-07 16:32:49', '2025-11-07 16:32:49', 0, '000000');
-INSERT INTO `t_workflow_component` VALUES (25, '0b4369bb60dc46d6bd84ceb4e36184dc', 'KeywordExtractor', '关键词提取', '从文本中提取关键词', 0, 1, '2025-12-26 16:30:05', '2025-12-26 16:30:05', 0, '000000');
 INSERT INTO `t_workflow_component` VALUES (26, 'bb00fc2f52c74fec82ee3f99725b56bb', 'Switcher', '条件分支', '根据条件执行不同分支', 0, 1, '2025-12-26 16:30:46', '2025-12-26 16:30:46', 0, '000000');
-INSERT INTO `t_workflow_component` VALUES (36, 'f37dbcb8f0d5464d90fbb22774490a56', 'HumanFeedback', '人类', '人机沟通', 0, 1, '2025-12-30 17:37:14', '2025-12-30 17:37:14', 0, '000000');
+INSERT INTO `t_workflow_component` VALUES (37, 'a7f8c2d44e5b4c83a9d6f103c2b47e18', 'Google', '网络搜索', '调用智谱 Web Search 检索互联网信息', 40, 1, '2026-07-29 20:30:00', '2026-07-29 20:30:00', 0, '000000');

 -- ----------------------------
 -- Table structure for t_workflow_edge
--- a/docs/script/sql/update/2026-07-21-sys-config-node-template.sql
+++ b/docs/script/sql/update/2026-07-21-sys-config-node-template.sql
@@ -1,5 +1,5 @@
 -- 补充工作流节点消息模板配置 (对应 issue IJX5VV)
-- 背景：NodeMessageTemplateEnum 依赖以下 9 个 sys_config 键，缺失时
+-- 背景：NodeMessageTemplateEnum 依赖以下 7 个 sys_config 键，缺失时
 --       WorkflowMessageUtil.getNodeMessageTemplate 会抛出「请先配置该节点的响应模板」。
 -- 这批配置在历史提交 20d531c0 中存在，SQL 脚本合并重命名时遗失，此处恢复。
 -- 幂等：按 config_key + tenant_id 判重，可重复执行。
@@ -20,10 +20,6 @@ INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`,
 SELECT 2027194134438277122, '000000', '结束节点响应模板', 'node.end.template', '🔚 流程已执行完毕，如果您有其他需求，请随时重新发起请求。', 'Y', 103, 1, '2026-02-27 09:28:40', 1, '2026-02-27 09:32:53', NULL
 FROM DUAL WHERE NOT EXISTS (SELECT 1 FROM `sys_config` WHERE `config_key` = 'node.end.template' AND `tenant_id` = '000000');

-INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`)
-SELECT 2027206492573335554, '000000', '人机交互节点响应模板', 'node.humanFeedback.template', '👤 人机交互节点：等待用户操作 - ', 'Y', 103, 1, '2026-02-27 10:17:46', 1, '2026-02-27 10:17:46', NULL
-FROM DUAL WHERE NOT EXISTS (SELECT 1 FROM `sys_config` WHERE `config_key` = 'node.humanFeedback.template' AND `tenant_id` = '000000');
-
 INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`)
 SELECT 2027208880369647617, '000000', '条件分支节点响应模板', 'node.switch.template', '🔀 条件分支节点：触发 -> 跳转到节点 ', 'Y', 103, 1, '2026-02-27 10:27:15', 1, '2026-02-27 10:35:54', NULL
 FROM DUAL WHERE NOT EXISTS (SELECT 1 FROM `sys_config` WHERE `config_key` = 'node.switch.template' AND `tenant_id` = '000000');
@@ -32,10 +28,6 @@ INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`,
 SELECT 2027213914603995137, '000000', '大模型回答节点响应模板', 'node.llmAnswer.template', '🤖 LLM 节点 生成回答：', 'Y', 103, 1, '2026-02-27 10:47:16', 1, '2026-02-27 10:52:40', NULL
 FROM DUAL WHERE NOT EXISTS (SELECT 1 FROM `sys_config` WHERE `config_key` = 'node.llmAnswer.template' AND `tenant_id` = '000000');

-INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`)
-SELECT 2027214387000066050, '000000', '关键词提取响应模板', 'node.keywordExtractor.template', '🔑 关键词提取节点 处理完成 ： ', 'Y', 103, 1, '2026-02-27 10:49:08', 1, '2026-02-27 10:52:08', NULL
-FROM DUAL WHERE NOT EXISTS (SELECT 1 FROM `sys_config` WHERE `config_key` = 'node.keywordExtractor.template' AND `tenant_id` = '000000');
-
 INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`)
 SELECT 2027217577397391361, '000000', '工作流异常响应模板', 'node.exception.template', '🛑 工作流发生异常：', 'N', 103, 1, '2026-02-27 11:01:49', 1, '2026-02-27 11:02:01', NULL
 FROM DUAL WHERE NOT EXISTS (SELECT 1 FROM `sys_config` WHERE `config_key` = 'node.exception.template' AND `tenant_id` = '000000');
--- a/docs/script/sql/update/2026-07-29-sys-config-node-template-all.sql
+++ b/docs/script/sql/update/2026-07-29-sys-config-node-template-all.sql
@@ -0,0 +1,39 @@
+-- 一次性补全工作流节点消息模板配置 (NodeMessageTemplateEnum 全部 8 个键)
+-- 背景：节点执行时 WorkflowMessageUtil.getNodeMessageTemplate 从 sys_config 读取展示模板,
+--       历史库中这批配置缺失, 导致运行工作流抛「请先配置该节点的响应模板」。
+--       其中前 7 个见 2026-07-21-sys-config-node-template.sql,
+--       Google Search 为此前从未入库的节点模板。
+-- 说明：代码已增加内置默认模板兜底, 本脚本为可选, 执行后模板可在 系统管理-配置管理 中自定义。
+-- 幂等：按 config_key + tenant_id 判重, 可重复执行。
+
+INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`)
+SELECT 2027192921483309058, '000000', 'HTTP请求节点响应模板', 'node.httpRequest.template', '✅ HTTP请求节点：结束响应 - ', 'Y', 103, 1, '2026-02-27 09:23:51', 1, '2026-02-27 09:31:41', NULL
+FROM DUAL WHERE NOT EXISTS (SELECT 1 FROM `sys_config` WHERE `config_key` = 'node.httpRequest.template' AND `tenant_id` = '000000');
+
+INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`)
+SELECT 2027193296990957569, '000000', '文生图节点响应模板', 'node.image.template', '🎨 文生图节点：结束响应 - 图片URL: ', 'Y', 103, 1, '2026-02-27 09:25:20', 1, '2026-02-27 09:31:52', NULL
+FROM DUAL WHERE NOT EXISTS (SELECT 1 FROM `sys_config` WHERE `config_key` = 'node.image.template' AND `tenant_id` = '000000');
+
+INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`)
+SELECT 2027193820393959425, '000000', '发送邮箱节点响应模板', 'node.mailsend.template', '📧 发送邮箱节点：结束响应 - ', 'Y', 103, 1, '2026-02-27 09:27:25', 1, '2026-02-27 09:32:05', NULL
+FROM DUAL WHERE NOT EXISTS (SELECT 1 FROM `sys_config` WHERE `config_key` = 'node.mailsend.template' AND `tenant_id` = '000000');
+
+INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`)
+SELECT 2027194134438277122, '000000', '结束节点响应模板', 'node.end.template', '🔚 流程已执行完毕，如果您有其他需求，请随时重新发起请求。', 'Y', 103, 1, '2026-02-27 09:28:40', 1, '2026-02-27 09:32:53', NULL
+FROM DUAL WHERE NOT EXISTS (SELECT 1 FROM `sys_config` WHERE `config_key` = 'node.end.template' AND `tenant_id` = '000000');
+
+INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`)
+SELECT 2027208880369647617, '000000', '条件分支节点响应模板', 'node.switch.template', '🔀 条件分支节点：触发 -> 跳转到节点 ', 'Y', 103, 1, '2026-02-27 10:27:15', 1, '2026-02-27 10:35:54', NULL
+FROM DUAL WHERE NOT EXISTS (SELECT 1 FROM `sys_config` WHERE `config_key` = 'node.switch.template' AND `tenant_id` = '000000');
+
+INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`)
+SELECT 2027213914603995137, '000000', '大模型回答节点响应模板', 'node.llmAnswer.template', '🤖 LLM 节点 生成回答：', 'Y', 103, 1, '2026-02-27 10:47:16', 1, '2026-02-27 10:52:40', NULL
+FROM DUAL WHERE NOT EXISTS (SELECT 1 FROM `sys_config` WHERE `config_key` = 'node.llmAnswer.template' AND `tenant_id` = '000000');
+
+INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`)
+SELECT 2027217577397391361, '000000', '工作流异常响应模板', 'node.exception.template', '🛑 工作流发生异常：', 'N', 103, 1, '2026-02-27 11:01:49', 1, '2026-02-27 11:02:01', NULL
+FROM DUAL WHERE NOT EXISTS (SELECT 1 FROM `sys_config` WHERE `config_key` = 'node.exception.template' AND `tenant_id` = '000000');
+
+INSERT INTO `sys_config` (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`, `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`, `update_time`, `remark`)
+SELECT 2084157200000000003, '000000', '网络搜索节点响应模板', 'node.googleSearch.template', '🔍 网络搜索节点处理完成：', 'Y', 103, 1, '2026-07-29 19:40:00', 1, '2026-07-29 19:40:00', NULL
+FROM DUAL WHERE NOT EXISTS (SELECT 1 FROM `sys_config` WHERE `config_key` = 'node.googleSearch.template' AND `tenant_id` = '000000');
--- a/docs/script/sql/update/2026-07-29-zhipu-web-search-node.sql
+++ b/docs/script/sql/update/2026-07-29-zhipu-web-search-node.sql
@@ -0,0 +1,66 @@
+-- 智谱 Web Search 工作流扩展节点
+-- 内部组件名继续使用 Google，以兼容现有前端组件和已保存流程。
+
+UPDATE `t_workflow_component`
+SET `title` = '网络搜索',
+    `remark` = '调用智谱 Web Search 检索互联网信息',
+    `display_order` = 40,
+    `is_enable` = 1,
+    `is_deleted` = 0,
+    `update_time` = NOW()
+WHERE `name` = 'Google'
+  AND `tenant_id` = '000000';
+
+INSERT INTO `t_workflow_component`
+    (`uuid`, `name`, `title`, `remark`, `display_order`, `is_enable`,
+     `create_time`, `update_time`, `is_deleted`, `tenant_id`)
+SELECT
+    'a7f8c2d44e5b4c83a9d6f103c2b47e18',
+    'Google',
+    '网络搜索',
+    '调用智谱 Web Search 检索互联网信息',
+    40,
+    1,
+    NOW(),
+    NOW(),
+    0,
+    '000000'
+FROM DUAL
+WHERE NOT EXISTS (
+    SELECT 1
+    FROM `t_workflow_component`
+    WHERE `name` = 'Google'
+      AND `tenant_id` = '000000'
+);
+
+UPDATE `sys_config`
+SET `config_name` = '网络搜索节点响应模板',
+    `config_value` = '🔍 网络搜索节点处理完成：',
+    `update_time` = NOW()
+WHERE `config_key` = 'node.googleSearch.template'
+  AND `tenant_id` = '000000';
+
+INSERT INTO `sys_config`
+    (`config_id`, `tenant_id`, `config_name`, `config_key`, `config_value`,
+     `config_type`, `create_dept`, `create_by`, `create_time`, `update_by`,
+     `update_time`, `remark`)
+SELECT
+    2084157200000000003,
+    '000000',
+    '网络搜索节点响应模板',
+    'node.googleSearch.template',
+    '🔍 网络搜索节点处理完成：',
+    'Y',
+    103,
+    1,
+    NOW(),
+    1,
+    NOW(),
+    '智谱 Web Search 工作流扩展节点'
+FROM DUAL
+WHERE NOT EXISTS (
+    SELECT 1
+    FROM `sys_config`
+    WHERE `config_key` = 'node.googleSearch.template'
+      AND `tenant_id` = '000000'
+);
--- a/pom.xml
+++ b/pom.xml
@@ -43,7 +43,6 @@
        <aws.sdk.version>2.28.22</aws.sdk.version>
        <!-- SMS 配置 -->
        <sms4j.version>3.3.5</sms4j.version>
-        <!-- FastJson已移除，使用Jackson替代 -->
        <!-- 面向运行时的D-ORM依赖 -->
        <anyline.version>8.7.2-20250603</anyline.version>
        <!-- 工作流配置 -->
@@ -62,6 +61,8 @@
        <weaviate.version>1.19.6</weaviate.version>
        <dify.version>1.2.7</dify.version>
        <coze.version>0.4.2</coze.version>
+        <!-- 智谱官方 Java SDK -->
+        <zai-sdk.version>0.3.5</zai-sdk.version>

        <!-- gRPC 版本 - 解决 Milvus SDK 依赖冲突 -->
        <grpc.version>1.62.2</grpc.version>
@@ -346,6 +347,12 @@
                <groupId>me.zhyd.oauth</groupId>
                <artifactId>JustAuth</artifactId>
                <version>${justauth.version}</version>
+                <exclusions>
+                    <exclusion>
+                        <groupId>com.alibaba</groupId>
+                        <artifactId>fastjson</artifactId>
+                    </exclusion>
+                </exclusions>
            </dependency>

            <!-- 离线IP地址定位库 ip2region -->
@@ -355,9 +362,6 @@
                <version>${ip2region.version}</version>
            </dependency>

-            <!-- FastJson已完全移除，项目统一使用Jackson -->
-            <!-- Jackson相关依赖已由Spring Boot统一管理 -->
-
            <dependency>
                <groupId>org.ruoyi</groupId>
                <artifactId>ruoyi-system</artifactId>
--- a/ruoyi-admin/src/main/resources/application-dev.yml
+++ b/ruoyi-admin/src/main/resources/application-dev.yml
@@ -60,7 +60,7 @@ spring:
          # rewriteBatchedStatements=true 批处理优化 大幅提升批量插入更新删除性能(对数据库有性能损耗 使用批量操作应考虑性能问题)
          url: jdbc:mysql://127.0.0.1:3306/ruoyi-ai?useUnicode=true&characterEncoding=utf8&zeroDateTimeBehavior=convertToNull&useSSL=true&serverTimezone=GMT%2B8&autoReconnect=true&rewriteBatchedStatements=true&allowPublicKeyRetrieval=true&nullCatalogMeansCurrent=true
          username: root
-          password: 123456
+          password: root
 #        agent:
 #          url: jdbc:mysql://127.0.0.1:3306/test?useUnicode=true&characterEncoding=utf8&zeroDateTimeBehavior=convertToNull&useSSL=true&serverTimezone=GMT%2B8&autoReconnect=true&rewriteBatchedStatements=true&allowPublicKeyRetrieval=true&nullCatalogMeansCurrent=true
 #          # url: jdbc:mysql://localhost:3306/agent_db
--- a/ruoyi-admin/src/main/resources/application.yml
+++ b/ruoyi-admin/src/main/resources/application.yml
@@ -332,6 +332,16 @@ vector-store:
    api-key:
    use-tls: false

+# 流程编排扩展节点
+workflow:
+  web-search:
+    zhipu:
+      # 推荐通过环境变量注入；为空时回退到模型管理中的 zhipu 厂商密钥
+      api-key: ${ZAI_API_KEY:}
+      base-url: ${ZHIPU_WEB_SEARCH_BASE_URL:https://open.bigmodel.cn/api/paas/v4/}
+      connect-timeout: ${ZHIPU_WEB_SEARCH_CONNECT_TIMEOUT:10}
+      read-timeout: ${ZHIPU_WEB_SEARCH_READ_TIMEOUT:30}
+
 # 短剧成片合成
 short-drama:
  composition:
--- a/ruoyi-common/ruoyi-common-chat/pom.xml
+++ b/ruoyi-common/ruoyi-common-chat/pom.xml
@@ -37,7 +37,6 @@
            <artifactId>ruoyi-common-sse</artifactId>
        </dependency>

-        <!-- FastJson已移除，使用Spring Boot自带的Jackson -->
        <dependency>
            <groupId>com.fasterxml.jackson.core</groupId>
            <artifactId>jackson-databind</artifactId>
--- a/ruoyi-common/ruoyi-common-chat/src/main/java/org/ruoyi/common/chat/enums/ErrorEnum.java
+++ b/ruoyi-common/ruoyi-common-chat/src/main/java/org/ruoyi/common/chat/enums/ErrorEnum.java
@@ -54,7 +54,6 @@ public enum ErrorEnum {
    A_WF_RUNTIME_NOT_FOUND("A00045", "工作流运行时数据找不到"),
    A_SEARCH_QUERY_IS_EMPTY("A00046", "搜索内容不能为空"),
    A_WF_COMPONENT_NOT_FOUND("A00047", "工作流基础组件找不到"),
-    A_WF_RESUME_FAIL("A00048", "工作流恢复执行时失败"),
    A_MAIL_SENDER_EMPTY("A00049", "邮件发送人不能为空"),
    A_MAIL_SENDER_CONFIG_ERROR("A00050", "邮件发送人配置错误"),
    A_MAIL_RECEIVER_EMPTY("A00051", "邮件接收人不能为空"),
--- a/ruoyi-common/ruoyi-common-chat/src/main/java/org/ruoyi/common/chat/service/workFlow/IWorkFlowStarterService.java
+++ b/ruoyi-common/ruoyi-common-chat/src/main/java/org/ruoyi/common/chat/service/workFlow/IWorkFlowStarterService.java
@@ -22,12 +22,4 @@ public interface IWorkFlowStarterService {
     * @return 流式输出结果
     */
    SseEmitter streaming(User user, String workflowUuid, List<ObjectNode> userInputs, Long sessionId);
-
-    /**
-     * 恢复工作流
-      * @param runtimeUuid 运行时UUID
-     * @param userInput 用户输入
-     * @param sseEmitter SSE连接对象
-     */
-    void resumeFlow(String runtimeUuid, String userInput, SseEmitter sseEmitter);
 }
--- a/ruoyi-common/ruoyi-common-social/src/main/java/me/zhyd/oauth/request/AbstractAuthWeChatEnterpriseRequest.java
+++ b/ruoyi-common/ruoyi-common-social/src/main/java/me/zhyd/oauth/request/AbstractAuthWeChatEnterpriseRequest.java
@@ -15,9 +15,6 @@ import me.zhyd.oauth.utils.HttpUtils;
 import me.zhyd.oauth.utils.StringUtils;
 import me.zhyd.oauth.utils.UrlBuilder;

-// 临时保留FastJson用于JustAuth库兼容
-import com.alibaba.fastjson.JSON;
-
 /**
 * <p>
 * 企业微信登录父类
@@ -64,11 +61,7 @@ public abstract class AbstractAuthWeChatEnterpriseRequest extends AuthDefaultReq
        String userTicket = object.has("user_ticket") ? object.get("user_ticket").asText() : null;
        JsonNode userDetail = getUserDetail(authToken.getAccessToken(), userId, userTicket);

-        // 将JsonNode转换为JSONObject以兼容JustAuth库
-        com.alibaba.fastjson.JSONObject rawUserInfo = com.alibaba.fastjson.JSON.parseObject(userDetail.toString());
-
        return AuthUser.builder()
-            .rawUserInfo(rawUserInfo)
            .username(userDetail.has("name") ? userDetail.get("name").asText() : null)
            .nickname(userDetail.has("alias") ? userDetail.get("alias").asText() : null)
            .avatar(userDetail.has("avatar") ? userDetail.get("avatar").asText() : null)
--- a/ruoyi-common/ruoyi-common-social/src/main/java/me/zhyd/oauth/request/AuthDingTalkV2Request.java
+++ b/ruoyi-common/ruoyi-common-social/src/main/java/me/zhyd/oauth/request/AuthDingTalkV2Request.java
@@ -19,9 +19,6 @@ import me.zhyd.oauth.utils.UrlBuilder;
 import java.util.HashMap;
 import java.util.Map;

-// 临时保留FastJson用于JustAuth库兼容
-import com.alibaba.fastjson.JSON;
-
 /**
 * 新版钉钉二维码登录
 *
@@ -92,13 +89,9 @@ public class AuthDingTalkV2Request extends AuthDefaultRequest {
            String response = new HttpUtils(config.getHttpConfig()).get(this.source.userInfo(), null, header, false).getBody();
            JsonNode object = objectMapper.readTree(response);

-            // 将JsonNode转换为JSONObject以兼容JustAuth库
-            com.alibaba.fastjson.JSONObject rawUserInfo = com.alibaba.fastjson.JSON.parseObject(object.toString());
-
            authToken.setOpenId(object.has("openId") ? object.get("openId").asText() : null);
            authToken.setUnionId(object.has("unionId") ? object.get("unionId").asText() : null);
            return AuthUser.builder()
-                .rawUserInfo(rawUserInfo)
                .uuid(object.has("unionId") ? object.get("unionId").asText() : null)
                .username(object.has("nick") ? object.get("nick").asText() : null)
                .nickname(object.has("nick") ? object.get("nick").asText() : null)
--- a/ruoyi-modules/ruoyi-aiflow/pom.xml
+++ b/ruoyi-modules/ruoyi-aiflow/pom.xml
@@ -24,6 +24,13 @@

    <dependencies>

+        <!-- 智谱官方 Java SDK：流程编排 Web Search 扩展节点 -->
+        <dependency>
+            <groupId>ai.z.openapi</groupId>
+            <artifactId>zai-sdk</artifactId>
+            <version>${zai-sdk.version}</version>
+        </dependency>
+
        <dependency>
            <groupId>org.ruoyi</groupId>
            <artifactId>ruoyi-common-chat</artifactId>
@@ -45,6 +52,11 @@
            <artifactId>ruoyi-common-satoken</artifactId>
        </dependency>

+        <dependency>
+            <groupId>org.ruoyi</groupId>
+            <artifactId>ruoyi-common-tenant</artifactId>
+        </dependency>
+
        <dependency>
            <groupId>org.ruoyi</groupId>
            <artifactId>ruoyi-common-mail</artifactId>
@@ -114,6 +126,7 @@
            <version>${langchain4j.version}</version>
            <scope>compile</scope>
        </dependency>
+
        <dependency>
            <groupId>org.jsoup</groupId>
            <artifactId>jsoup</artifactId>
@@ -133,8 +146,6 @@
            <version>${langgraph4j.version}</version>
        </dependency>

-
-
        <dependency>
            <groupId>org.apache.commons</groupId>
            <artifactId>commons-collections4</artifactId>
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/controller/WorkflowRuntimeController.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/controller/WorkflowRuntimeController.java
@@ -1,16 +1,13 @@
 package org.ruoyi.workflow.controller;

 import com.baomidou.mybatisplus.extension.plugins.pagination.Page;
-import io.swagger.v3.oas.annotations.Operation;
 import jakarta.annotation.Resource;
 import jakarta.validation.constraints.Min;
 import jakarta.validation.constraints.NotNull;
 import org.ruoyi.common.core.domain.R;
 import org.ruoyi.workflow.dto.workflow.WfRuntimeNodeDto;
 import org.ruoyi.workflow.dto.workflow.WfRuntimeResp;
-import org.ruoyi.workflow.dto.workflow.WorkflowResumeReq;
 import org.ruoyi.workflow.service.WorkflowRuntimeService;
-import org.ruoyi.workflow.workflow.WorkflowStarter;
 import org.springframework.validation.annotation.Validated;
 import org.springframework.web.bind.annotation.*;

@@ -24,16 +21,6 @@ public class WorkflowRuntimeController {
    @Resource
    private WorkflowRuntimeService workflowRuntimeService;

-    @Resource
-    private WorkflowStarter workflowStarter;
-
-    @Operation(summary = "接收用户输入以继续执行剩余流程")
-    @PostMapping(value = "/resume/{runtimeUuid}")
-    public R resume(@PathVariable String runtimeUuid, @RequestBody WorkflowResumeReq resumeReq) {
-        workflowStarter.resumeFlow(runtimeUuid, resumeReq.getFeedbackContent(), resumeReq.getSseEmitter());
-        return R.ok();
-    }
-
    @GetMapping("/page")
    public R<Page<WfRuntimeResp>> search(@RequestParam String wfUuid,
                                         @NotNull @Min(1) Integer currentPage,
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/cosntant/AdiConstant.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/cosntant/AdiConstant.java
@@ -337,7 +337,6 @@ public class AdiConstant {
        public static final String DEFAULT_INPUT_PARAM_NAME = "input";
        public static final String DEFAULT_OUTPUT_PARAM_NAME = "output";
        public static final String DEFAULT_ERROR_OUTPUT_PARAM_NAME = "error_msg";
-        public static final String HUMAN_FEEDBACK_KEY = "human_feedback";
        public static final int NODE_PROCESS_STATUS_READY = 1;
        public static final int NODE_PROCESS_STATUS_DOING = 2;
        public static final int NODE_PROCESS_STATUS_SUCCESS = 3;
@@ -347,7 +346,6 @@ public class AdiConstant {
        public static final int WORKFLOW_PROCESS_STATUS_DOING = 2;
        public static final int WORKFLOW_PROCESS_STATUS_SUCCESS = 3;
        public static final int WORKFLOW_PROCESS_STATUS_FAIL = 4;
-        public static final int WORKFLOW_PROCESS_STATUS_WAITING_INPUT = 5;

        public static final int WORKFLOW_NODE_PROCESS_TYPE_NORMAL = 1;
        public static final int WORKFLOW_NODE_PROCESS_TYPE_CONDITIONAL = 2;
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/dto/workflow/WorkflowResumeReq.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/dto/workflow/WorkflowResumeReq.java
@@ -1,10 +0,0 @@
-package org.ruoyi.workflow.dto.workflow;
-
-import lombok.Data;
-import org.springframework.web.servlet.mvc.method.annotation.SseEmitter;
-
-@Data
-public class WorkflowResumeReq {
-    private String feedbackContent;
-    private SseEmitter sseEmitter;
-}
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/helper/SSEEmitterHelper.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/helper/SSEEmitterHelper.java
@@ -55,8 +55,13 @@ public class SSEEmitterHelper {
            } else {
                sseEmitter.send(msg);
            }
+        } catch (IllegalStateException ise) {
+            // SSE连接已关闭（用户刷新页面、关闭标签页或重新提交）
+            log.warn("SSE emitter already completed for event [{}], ignoring", name);
+            COMPLETED_SSE.put(sseEmitter, Boolean.TRUE);
        } catch (IOException ioException) {
            log.error("stream onNext error", ioException);
+            COMPLETED_SSE.put(sseEmitter, Boolean.TRUE);
        }
    }

--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/util/WorkflowMessageUtil.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/util/WorkflowMessageUtil.java
@@ -4,7 +4,6 @@ import org.ruoyi.common.chat.domain.dto.request.ChatRequest;
 import org.ruoyi.common.chat.domain.vo.chat.ChatModelVo;
 import org.ruoyi.common.chat.enums.RoleType;
 import lombok.extern.slf4j.Slf4j;
-import org.ruoyi.common.core.exception.ServiceException;
 import org.ruoyi.common.core.service.ConfigService;
 import org.ruoyi.common.core.utils.SpringUtils;
 import org.ruoyi.common.core.utils.StringUtils;
@@ -12,6 +11,7 @@ import org.ruoyi.workflow.entity.WorkflowNode;
 import org.ruoyi.workflow.helper.SSEEmitterHelper;
 import org.ruoyi.workflow.workflow.WfState;
 import org.ruoyi.workflow.workflow.WorkflowUtil;
+import org.ruoyi.workflow.workflow.node.enmus.NodeMessageTemplateEnum;
 import org.springframework.web.servlet.mvc.method.annotation.SseEmitter;

 /**
@@ -38,17 +38,20 @@ public class WorkflowMessageUtil {


    /**
-     * 获取节点的响应模板
+     * 获取节点的响应模板 <br/>
+     * 优先读取 sys_config 配置(可在系统管理-配置管理中自定义),
+     * 未配置时回退到枚举内置默认模板, 模板仅为展示文案, 缺失不应中断工作流执行
     * @param configKey 参数Key
     * @return 返回模板样式
     */
    public static String getNodeMessageTemplate(String configKey){
        ConfigService configService = SpringUtil.getBean(ConfigService.class);
        String configValue = configService.getConfigValue(configKey);
-        if (StringUtils.isEmpty(configValue)) {
-            throw new ServiceException("请先配置该节点的响应模板");
+        if (StringUtils.isNotEmpty(configValue)) {
+            return configValue;
        }
-        return configValue;
+        log.warn("sys_config 未配置节点响应模板 [{}], 已回退使用内置默认模板", configKey);
+        return NodeMessageTemplateEnum.getDefaultTemplate(configKey);
    }

    /**
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/InterruptedFlow.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/InterruptedFlow.java
@@ -1,17 +0,0 @@
-package org.ruoyi.workflow.workflow;
-
-import org.apache.commons.collections4.map.PassiveExpiringMap;
-
-/**
- * 已中断正在等待用户输入的流程 <br/>
- * TODO 需要考虑项目多节点部署的情况
- */
-public class InterruptedFlow {
-
-    /**
-     * 10分钟超时
-     */
-    private static final PassiveExpiringMap.ExpirationPolicy<String, WorkflowEngine> ep = new PassiveExpiringMap.ConstantTimeToLiveExpirationPolicy<>(60 * 1000 * 10);
-    public static PassiveExpiringMap<String, WorkflowEngine> RUNTIME_TO_GRAPH = new PassiveExpiringMap<>(ep);
-
-}
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/WfComponentNameEnum.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/WfComponentNameEnum.java
@@ -16,24 +16,14 @@ public enum WfComponentNameEnum {

    TONGYI_WANX("Tongyiwanx"),

-    DOCUMENT_EXTRACTOR("DocumentExtractor"),
-
-    KEYWORD_EXTRACTOR("KeywordExtractor"),
-
    FAQ_EXTRACTOR("FaqExtractor"),

    KNOWLEDGE_RETRIEVER("KnowledgeRetrieval"),

    SWITCHER("Switcher"),

-    CLASSIFIER("Classifier"),
-
-    TEMPLATE("Template"),
-
    GOOGLE_SEARCH("Google"),

-    HUMAN_FEEDBACK("HumanFeedback"),
-
    MAIL_SEND("MailSend"),

    HTTP_REQUEST("HttpRequest");
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/WfNodeFactory.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/WfNodeFactory.java
@@ -4,15 +4,14 @@ import org.ruoyi.workflow.entity.WorkflowComponent;
 import org.ruoyi.workflow.entity.WorkflowNode;
 import org.ruoyi.workflow.workflow.node.AbstractWfNode;
 import org.ruoyi.workflow.workflow.node.EndNode;
-import org.ruoyi.workflow.workflow.node.humanFeedBack.HumanFeedbackNode;
 import org.ruoyi.workflow.workflow.node.answer.LLMAnswerNode;
 import org.ruoyi.workflow.workflow.node.httpRequest.HttpRequestNode;
 import org.ruoyi.workflow.workflow.node.image.ImageNode;
-import org.ruoyi.workflow.workflow.node.keywordExtractor.KeywordExtractorNode;
 import org.ruoyi.workflow.workflow.node.knowledgeRetrieval.KnowledgeRetrievalNode;
 import org.ruoyi.workflow.workflow.node.mailSend.MailSendNode;
 import org.ruoyi.workflow.workflow.node.start.StartNode;
 import org.ruoyi.workflow.workflow.node.switcher.SwitcherNode;
+import org.ruoyi.workflow.workflow.node.googleSearch.GoogleSearchNode;

 public class WfNodeFactory {
    public static AbstractWfNode create(WorkflowComponent wfComponent, WorkflowNode nodeDefinition,
@@ -21,14 +20,13 @@ public class WfNodeFactory {
        switch (WfComponentNameEnum.getByName(wfComponent.getName())) {
            case START -> wfNode = new StartNode(wfComponent, nodeDefinition, wfState, nodeState);
            case LLM_ANSWER -> wfNode = new LLMAnswerNode(wfComponent, nodeDefinition, wfState, nodeState);
-            case KEYWORD_EXTRACTOR -> wfNode = new KeywordExtractorNode(wfComponent, nodeDefinition, wfState, nodeState);
            case TONGYI_WANX -> wfNode = new ImageNode(wfComponent, nodeDefinition, wfState, nodeState);
            case KNOWLEDGE_RETRIEVER -> wfNode = new KnowledgeRetrievalNode(wfComponent, nodeDefinition, wfState, nodeState);
            case END -> wfNode = new EndNode(wfComponent, nodeDefinition, wfState, nodeState);
            case MAIL_SEND -> wfNode = new MailSendNode(wfComponent, nodeDefinition, wfState, nodeState);
            case HTTP_REQUEST -> wfNode = new HttpRequestNode(wfComponent, nodeDefinition, wfState, nodeState);
            case SWITCHER -> wfNode = new SwitcherNode(wfComponent, nodeDefinition, wfState, nodeState);
-            case HUMAN_FEEDBACK -> wfNode = new HumanFeedbackNode(wfComponent, nodeDefinition, wfState, nodeState);
+            case GOOGLE_SEARCH -> wfNode = new GoogleSearchNode(wfComponent, nodeDefinition, wfState, nodeState);
            default -> {
            }
        }
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/WfState.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/WfState.java
@@ -55,11 +55,6 @@ public class WfState {
    private List<NodeIOData> output = new ArrayList<>();
    private Integer processStatus = WORKFLOW_PROCESS_STATUS_READY;

-    /**
-     * 人机交互节点
-     */
-    private Set<String> interruptNodes = new HashSet<>();
-
    public WfState(User user, List<NodeIOData> input, String uuid, Long userId, String tokenValue, SseEmitter sseEmitter, Long sessionId) {
        this.input = input;
        this.user = user;
@@ -133,8 +128,4 @@ public class WfState {
                .findFirst()
                .orElse(null);
    }
-
-    public void addInterruptNode(String nodeUuid) {
-        this.interruptNodes.add(nodeUuid);
-    }
 }
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/WorkflowEngine.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/WorkflowEngine.java
@@ -108,82 +108,39 @@ public class WorkflowEngine {

            MemorySaver saver = new MemorySaver();
            CompileConfig compileConfig = CompileConfig.builder().checkpointSaver(saver)
-                    .interruptBefore(wfState.getInterruptNodes().toArray(String[]::new))
                    .build();
            app = mainStateGraph.compile(compileConfig);
            RunnableConfig invokeConfig = RunnableConfig.builder().build();
-            exe(invokeConfig, false);
+            exe(invokeConfig);
        } catch (Exception e) {
            errorWhenExe(e);
        }
    }

-    private void exe(RunnableConfig invokeConfig, boolean resume) {
+    private void exe(RunnableConfig invokeConfig) {
        //不使用langgraph4j state的update相关方法，无需传入input
-        AsyncGenerator<NodeOutput<WfNodeState>> outputs = app.stream(resume ? null : Map.of(), invokeConfig);
+        AsyncGenerator<NodeOutput<WfNodeState>> outputs = app.stream(Map.of(), invokeConfig);
        streamingResult(wfState, outputs, sseEmitter);

        StateSnapshot<WfNodeState> stateSnapshot = app.getState(invokeConfig);
-        String nextNode = stateSnapshot.config().nextNode().orElse("");
-        //还有下个节点，表示进入中断状态，等待用户输入后继续执<E7BBAD>?
-        if (StringUtils.isNotBlank(nextNode) && !nextNode.equalsIgnoreCase(END)) {
-            // 获取提示模板
-            String nodeMessageTemplate = WorkflowMessageUtil.getNodeMessageTemplate(NodeMessageTemplateEnum.HUMAN_FEED_BACK.getValue());
-            // 获取人机交互提示信息
-            String intTip = nodeMessageTemplate + WorkflowUtil.getHumanFeedbackTip(nextNode, wfNodes);
-            //将等待输入信息[事件与提示词]发送到到客户端
-            SSEEmitterHelper.parseAndSendPartialMsg(sseEmitter, "[NODE_WAIT_FEEDBACK_BY_" + nextNode + "]", intTip);
-            // 保存提示信息到Chat信息记录中（对话使用）
-            WorkflowMessageUtil.saveWorkflowMessage(wfState, intTip);
-            InterruptedFlow.RUNTIME_TO_GRAPH.put(wfState.getUuid(), this);
-            //更新状<E696B0>?
-            wfState.setProcessStatus(WORKFLOW_PROCESS_STATUS_WAITING_INPUT);
-            workflowRuntimeService.updateOutput(wfRuntimeResp.getId(), wfState);
-        } else {
-            WorkflowRuntime updatedRuntime = workflowRuntimeService.updateOutput(wfRuntimeResp.getId(), wfState);
-            // 保存成功会话信息
-            wfNodes.stream().filter(item -> stateSnapshot.node().equals(item.getUuid()))
-                .findFirst().ifPresent(wfNode -> {
-                    // 获取节点模板提示词信息
-                    String nodeMessageTemplate = WorkflowMessageUtil.getNodeMessageTemplate(NodeMessageTemplateEnum.END.getValue());
-                    // 发送SSE消息驱动事件和保存会话
-                    WorkflowMessageUtil.notifyAndStoreMessage(wfState, sseEmitter, wfNode, nodeMessageTemplate);
-            });
-            // 发送结束消息
-            sseEmitterHelper.sendComplete(user.getId(), sseEmitter, updatedRuntime.getOutput());
-            // 发送驱动消息事件
-            InterruptedFlow.RUNTIME_TO_GRAPH.remove(wfState.getUuid());
-        }
-    }
-
-    /**
-     * 中断流程等待用户输入时，会进行暂停状态，用户输入后调用本方法执行流程剩余部分
-     *
-     * @param userInput 用户输入
-     */
-    public void resume(String userInput) {
-        RunnableConfig invokeConfig = RunnableConfig.builder().build();
-        try {
-            app.updateState(invokeConfig, Map.of(HUMAN_FEEDBACK_KEY, userInput), null);
-            exe(invokeConfig, true);
-        } catch (Exception e) {
-            errorWhenExe(e);
-        } finally {
-            //有可能多次接收人机交互，待整个流程完全执行后才能删除
-            if (wfState.getProcessStatus() != WORKFLOW_PROCESS_STATUS_WAITING_INPUT) {
-                InterruptedFlow.RUNTIME_TO_GRAPH.remove(wfState.getUuid());
-            }
-        }
+        wfState.setProcessStatus(WORKFLOW_PROCESS_STATUS_SUCCESS);
+        WorkflowRuntime updatedRuntime = workflowRuntimeService.updateOutput(wfRuntimeResp.getId(), wfState);
+        wfNodes.stream().filter(item -> stateSnapshot.node().equals(item.getUuid()))
+            .findFirst().ifPresent(wfNode -> {
+                String nodeMessageTemplate = WorkflowMessageUtil.getNodeMessageTemplate(NodeMessageTemplateEnum.END.getValue());
+                WorkflowMessageUtil.notifyAndStoreMessage(wfState, sseEmitter, wfNode, nodeMessageTemplate);
+        });
+        sseEmitterHelper.sendComplete(user.getId(), sseEmitter, updatedRuntime.getOutput());
    }

    private void errorWhenExe(Exception e) {
        log.error("error", e);
        String nodeMessageTemplate = WorkflowMessageUtil.getNodeMessageTemplate(NodeMessageTemplateEnum.EXCEPTION.getValue());
        String errorMsg = e.getMessage();
-        if (errorMsg.contains("parallel node doesn't support conditional branch")) {
+        if (errorMsg != null && errorMsg.contains("parallel node doesn't support conditional branch")) {
            errorMsg = "并行节点中不能包含条件分<EFBFBD>?";
        }
-        errorMsg = nodeMessageTemplate + errorMsg;
+        errorMsg = nodeMessageTemplate + (errorMsg != null ? errorMsg : e.getClass().getSimpleName());
        // 保存会话信息且发送驱动消息事件
        WorkflowMessageUtil.saveWorkflowMessage(wfState, errorMsg);
        sseEmitterHelper.sendErrorAndComplete(user.getId(), sseEmitter, errorMsg);
@@ -267,6 +224,10 @@ public class WorkflowEngine {
                log.info("node:{},chunk:{}", node, chunk);
                SSEEmitterHelper.parseAndSendPartialMsg(sseEmitter, "[NODE_CHUNK_" + node + "]", chunk);
            } else {
+                // __END__ 是 langgraph4j 的终止伪节点, 无对应业务节点状态, 跳过
+                if (END.equals(out.node())) {
+                    continue;
+                }
                AbstractWfNode abstractWfNode = wfState.getCompletedNodes().stream()
                        .filter(item -> item.getNode().getUuid().endsWith(out.node())).findFirst().orElse(null);
                if (null != abstractWfNode) {
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/WorkflowGraphBuilder.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/WorkflowGraphBuilder.java
@@ -19,7 +19,6 @@ import static org.bsc.langgraph4j.StateGraph.END;
 import static org.bsc.langgraph4j.StateGraph.START;
 import static org.bsc.langgraph4j.action.AsyncEdgeAction.edge_async;
 import static org.bsc.langgraph4j.action.AsyncNodeAction.node_async;
-import static org.ruoyi.workflow.workflow.WfComponentNameEnum.HUMAN_FEEDBACK;

 /**
 * 负责构建工作流运行所依赖的状态图<E68081>?
@@ -27,7 +26,6 @@ import static org.ruoyi.workflow.workflow.WfComponentNameEnum.HUMAN_FEEDBACK;
@Slf4j
 public class WorkflowGraphBuilder {

-    private final Map<Long, WorkflowComponent> componentIndex;
    private final Map<String, WorkflowNode> nodeIndex;
    private final Map<String, List<WorkflowEdge>> edgesBySource;
    private final Map<String, List<WorkflowEdge>> edgesByTarget;
@@ -46,8 +44,6 @@ public class WorkflowGraphBuilder {
            List<WorkflowEdge> edges,
            WorkflowNodeRunner nodeRunner,
            WfState wfState) {
-        this.componentIndex = components.stream()
-                .collect(Collectors.toMap(WorkflowComponent::getId, Function.identity(), (origin, ignore) -> origin));
        this.nodeIndex = nodes.stream()
                .collect(Collectors.toMap(WorkflowNode::getUuid, Function.identity(), (origin, ignore) -> origin));
        this.edgesBySource = edges.stream().collect(Collectors.groupingBy(WorkflowEdge::getSourceNodeUuid));
@@ -217,14 +213,6 @@ public class WorkflowGraphBuilder {
        WorkflowNode wfNode = getNodeByUuid(stateGraphNodeUuid);
        stateGraph.addNode(stateGraphNodeUuid, node_async(state -> nodeRunner.run(wfNode, state)));
        stateGraphList.add(stateGraph);
-
-        WorkflowComponent component = componentIndex.get(wfNode.getWorkflowComponentId());
-        if (component == null) {
-            throw new BaseException(ErrorEnum.A_PARAMS_ERROR.getInfo());
-        }
-        if (HUMAN_FEEDBACK.getName().equals(component.getName())) {
-            wfState.addInterruptNode(stateGraphNodeUuid);
-        }
    }

    private void addEdgeToStateGraph(StateGraph<WfNodeState> stateGraph, String source, String target) throws GraphStateException {
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/WorkflowStarter.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/WorkflowStarter.java
@@ -6,9 +6,9 @@ import jakarta.annotation.Resource;
 import lombok.extern.slf4j.Slf4j;
 import org.ruoyi.common.chat.entity.User;
 import org.ruoyi.common.chat.service.workFlow.IWorkFlowStarterService;
-import org.ruoyi.common.core.exception.base.BaseException;
 import org.ruoyi.common.satoken.utils.LoginHelper;
 import org.ruoyi.common.sse.core.SseEmitterManager;
+import org.ruoyi.common.tenant.helper.TenantHelper;
 import org.ruoyi.workflow.entity.*;
 import org.ruoyi.workflow.helper.SSEEmitterHelper;
 import org.ruoyi.workflow.service.*;
@@ -58,6 +58,8 @@ public class WorkflowStarter implements IWorkFlowStarterService {
        Long userId = LoginHelper.getUserId();
        // 获取登录Token（仅透传给 WfState，工作流 SSE 通过 emitter 直发，不串台）
        String tokenValue = StpUtil.getTokenValue();
+        // 获取当前租户ID（@Async 线程不继承请求线程的租户上下文，需显式透传）
+        String tenantId = TenantHelper.getTenantId();
        // 根据会话ID连接SSE对象（每会话一个连接，避免同用户多会话串台）
        SseEmitter sseEmitter = sseEmitterManager.connect(String.valueOf(sessionId));
        if (!sseEmitterHelper.checkOrComplete(user, sseEmitter)) {
@@ -71,41 +73,33 @@ public class WorkflowStarter implements IWorkFlowStarterService {
            sseEmitterHelper.sendErrorAndComplete(user.getId(), sseEmitter, A_WF_DISABLED.getInfo());
            return sseEmitter;
        }
-        self.asyncRun(user, workflow, userInputs, sseEmitter, userId, tokenValue, sessionId);
+        self.asyncRun(user, workflow, userInputs, sseEmitter, userId, tokenValue, sessionId, tenantId);
        return sseEmitter;
    }

    @Async
-    public void asyncRun(User user, Workflow workflow, List<ObjectNode> userInputs, SseEmitter sseEmitter, Long userId, String tokenValue, Long sessionId) {
-        log.info("WorkflowEngine run,userId:{},workflowUuid:{},userInputs:{}", user.getId(), workflow.getUuid(), userInputs);
-        List<WorkflowComponent> components = workflowComponentService.getAllEnable();
-        List<WorkflowNode> nodes = workflowNodeService.lambdaQuery()
-                .eq(WorkflowNode::getWorkflowId, workflow.getId())
-                .eq(WorkflowNode::getIsDeleted, false)
-                .list();
-        List<WorkflowEdge> edges = workflowEdgeService.lambdaQuery()
-                .eq(WorkflowEdge::getWorkflowId, workflow.getId())
-                .eq(WorkflowEdge::getIsDeleted, false)
-                .list();
-        WorkflowEngine workflowEngine = new WorkflowEngine(workflow,
-                sseEmitterHelper, components, nodes, edges,
-                workflowRuntimeService, workflowRuntimeNodeService);
-        workflowEngine.run(user, userInputs, sseEmitter, userId, tokenValue, sessionId);
-    }
-
-    @Async
-    public void resumeFlow(String runtimeUuid, String userInput, SseEmitter sseEmitter) {
-        WorkflowEngine workflowEngine = InterruptedFlow.RUNTIME_TO_GRAPH.get(runtimeUuid);
-        if (null == workflowEngine) {
-            log.error("工作流恢复执行时失败,runtime:{}", runtimeUuid);
-            throw new BaseException(A_WF_RESUME_FAIL.getInfo());
+    public void asyncRun(User user, Workflow workflow, List<ObjectNode> userInputs, SseEmitter sseEmitter, Long userId, String tokenValue, Long sessionId, String tenantId) {
+        // @Async 线程不继承请求线程的租户上下文, 显式设置, 避免租户缓存/隔离逻辑异常
+        if (tenantId != null) {
+            TenantHelper.setDynamic(tenantId);
        }
-        // 如果SSE连接对象不为空传入该对象（Chat调用工作流对话使用）
-        if (null != sseEmitter){
-            workflowEngine.setSseEmitter(sseEmitter);
-            // 为了让每个节点都可以发送模板消息 保持SSE对象一致（以防出现向已关闭的SSE对象发送消息）
-            workflowEngine.getWfState().setSseEmitter(sseEmitter);
+        try {
+            log.info("WorkflowEngine run,userId:{},workflowUuid:{},userInputs:{}", user.getId(), workflow.getUuid(), userInputs);
+            List<WorkflowComponent> components = workflowComponentService.getAllEnable();
+            List<WorkflowNode> nodes = workflowNodeService.lambdaQuery()
+                    .eq(WorkflowNode::getWorkflowId, workflow.getId())
+                    .eq(WorkflowNode::getIsDeleted, false)
+                    .list();
+            List<WorkflowEdge> edges = workflowEdgeService.lambdaQuery()
+                    .eq(WorkflowEdge::getWorkflowId, workflow.getId())
+                    .eq(WorkflowEdge::getIsDeleted, false)
+                    .list();
+            WorkflowEngine workflowEngine = new WorkflowEngine(workflow,
+                    sseEmitterHelper, components, nodes, edges,
+                    workflowRuntimeService, workflowRuntimeNodeService);
+            workflowEngine.run(user, userInputs, sseEmitter, userId, tokenValue, sessionId);
+        } finally {
+            TenantHelper.clearDynamic();
        }
-        workflowEngine.resume(userInput);
    }
 }
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/WorkflowUtil.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/WorkflowUtil.java
@@ -2,7 +2,6 @@ package org.ruoyi.workflow.workflow;

 import cn.hutool.core.collection.CollStreamUtil;
 import cn.hutool.core.collection.CollUtil;
-import cn.hutool.core.util.StrUtil;
 import dev.langchain4j.data.message.ChatMessage;
 import dev.langchain4j.data.message.UserMessage;
 import dev.langchain4j.model.chat.response.StreamingChatResponseHandler;
@@ -20,7 +19,6 @@ import org.ruoyi.common.chat.factory.ImageServiceFactory;
 import org.ruoyi.workflow.base.NodeInputConfigTypeHandler;
 import org.ruoyi.workflow.entity.WorkflowNode;
 import org.ruoyi.workflow.enums.WfIODataTypeEnum;
-import org.ruoyi.workflow.util.JsonUtil;
 import org.ruoyi.workflow.workflow.data.NodeIOData;
 import org.ruoyi.workflow.workflow.data.NodeIODataContent;
 import org.ruoyi.workflow.workflow.def.WfNodeParamRef;
@@ -88,22 +86,6 @@ public class WorkflowUtil{
        return result;
    }

-    public static String getHumanFeedbackTip(String nodeUuid, List<WorkflowNode> wfNodes) {
-        WorkflowNode wfNode = wfNodes.stream()
-            .filter(item -> item.getUuid().equals(nodeUuid))
-            .findFirst().orElse(null);
-        if (null == wfNode) {
-            return "";
-        }
-        String wfNodeNodeConfig = wfNode.getNodeConfig();
-        if (StrUtil.isBlank(wfNodeNodeConfig)) {
-            return "";
-        }
-        Map<String, Object> map = JsonUtil.toMap(wfNodeNodeConfig);
-        Object tip = map.getOrDefault("tip", "");
-        return String.valueOf(tip);
-    }
-
    public void streamingInvokeLLM(WfState wfState, WfNodeState state, WorkflowNode node, String modelName,
                                   String prompt, String nodeMessageTemplate) {
        log.info("stream invoke, modelName: {}", modelName);
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/classifier/ClassifierNodeConfig.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/classifier/ClassifierNodeConfig.java
@@ -1,16 +0,0 @@
-package org.ruoyi.workflow.workflow.node.classifier;
-
-import com.fasterxml.jackson.annotation.JsonProperty;
-import lombok.Data;
-
-import java.util.ArrayList;
-import java.util.List;
-
-@Data
-public class ClassifierNodeConfig {
-    private List categories = new ArrayList<>();
-    @JsonProperty("model_platform")
-    private String modelPlatform;
-    @JsonProperty("model_name")
-    private String modelName;
-}
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/enmus/NodeMessageTemplateEnum.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/enmus/NodeMessageTemplateEnum.java
@@ -3,23 +3,44 @@ package org.ruoyi.workflow.workflow.node.enmus;
 import lombok.Getter;

 /**
- * 节点消息模板ConfigKey枚举
+ * 节点消息模板ConfigKey枚举 <br/>
+ * 模板优先从 sys_config 读取(可在系统管理-配置管理中自定义), 未配置时回退到 defaultTemplate 内置默认值
 */
@Getter
 public enum NodeMessageTemplateEnum {
-    HTTP_REQUEST("node.httpRequest.template"),
-    MAIL_SEND("node.mailsend.template"),
-    IMAGE("node.image.template"),
-    HUMAN_FEED_BACK("node.humanFeedback.template"),
-    SWITCH("node.switch.template"),
-    LLM_RESPONSE("node.llmAnswer.template"),
-    KEYWORD_EXTRACTOR("node.keywordExtractor.template"),
-    EXCEPTION("node.exception.template"),
-    END("node.end.template");
+    HTTP_REQUEST("node.httpRequest.template", "✅ HTTP请求节点：结束响应 - "),
+    MAIL_SEND("node.mailsend.template", "📧 发送邮箱节点：结束响应 - "),
+    IMAGE("node.image.template", "🎨 文生图节点：结束响应 - 图片URL: "),
+    SWITCH("node.switch.template", "🔀 条件分支节点：触发 -> 跳转到节点 "),
+    LLM_RESPONSE("node.llmAnswer.template", "🤖 LLM 节点 生成回答："),
+    GOOGLE_SEARCH("node.googleSearch.template", "🔍 网络搜索节点处理完成："),
+    EXCEPTION("node.exception.template", "🛑 工作流发生异常："),
+    END("node.end.template", "🔚 流程已执行完毕，如果您有其他需求，请随时重新发起请求。");

    private final String value;

-    NodeMessageTemplateEnum(String value) {
+    /**
+     * 内置默认模板, sys_config 未配置对应键时使用
+     */
+    private final String defaultTemplate;
+
+    NodeMessageTemplateEnum(String value, String defaultTemplate) {
        this.value = value;
+        this.defaultTemplate = defaultTemplate;
+    }
+
+    /**
+     * 根据 configKey 获取内置默认模板, 未知键返回空串
+     *
+     * @param configKey sys_config 配置键
+     * @return 内置默认模板
+     */
+    public static String getDefaultTemplate(String configKey) {
+        for (NodeMessageTemplateEnum item : values()) {
+            if (item.value.equals(configKey)) {
+                return item.defaultTemplate;
+            }
+        }
+        return "";
    }
 }
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/googleSearch/GoogleSearchNode.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/googleSearch/GoogleSearchNode.java
@@ -0,0 +1,91 @@
+package org.ruoyi.workflow.workflow.node.googleSearch;
+
+import lombok.extern.slf4j.Slf4j;
+import org.apache.commons.lang3.StringUtils;
+import org.ruoyi.workflow.entity.WorkflowComponent;
+import org.ruoyi.workflow.entity.WorkflowNode;
+import org.ruoyi.workflow.util.JsonUtil;
+import org.ruoyi.workflow.util.SpringUtil;
+import org.ruoyi.workflow.workflow.NodeProcessResult;
+import org.ruoyi.workflow.workflow.WfNodeState;
+import org.ruoyi.workflow.workflow.WfState;
+import org.ruoyi.workflow.workflow.WorkflowUtil;
+import org.ruoyi.workflow.workflow.data.NodeIOData;
+import org.ruoyi.workflow.workflow.node.AbstractWfNode;
+import org.ruoyi.workflow.workflow.node.enmus.NodeMessageTemplateEnum;
+
+import java.util.List;
+import java.util.UUID;
+
+import static org.ruoyi.workflow.cosntant.AdiConstant.WorkflowConstant.DEFAULT_OUTPUT_PARAM_NAME;
+
+/**
+ * 【扩展节点】网络搜索
+ * 通过智谱 Web Search API 返回适合大模型消费的结构化网页结果。
+ */
+@Slf4j
+public class GoogleSearchNode extends AbstractWfNode {
+
+    public GoogleSearchNode(WorkflowComponent wfComponent, WorkflowNode nodeDef, WfState wfState, WfNodeState nodeState) {
+        super(wfComponent, nodeDef, wfState, nodeState);
+    }
+
+    /**
+     * 处理搜索请求
+     * nodeConfig 格式：
+     * {
+     *   "query": "搜索关键词",
+     *   "search_engine": "search_std",
+     *   "result_count": 10,
+     *   "search_domain_filter": "",
+     *   "search_recency_filter": "noLimit",
+     *   "content_size": "medium",
+     *   "include_image": false
+     * }
+     *
+     * @return 搜索结果
+     */
+    @Override
+    public NodeProcessResult onProcess() {
+        GoogleSearchNodeConfig config = checkAndGetConfig(GoogleSearchNodeConfig.class);
+
+        // 获取搜索关键词
+        String searchQuery = WorkflowUtil.renderTemplate(config.getQuery(), state.getInputs());
+        if (StringUtils.isBlank(searchQuery)) {
+            searchQuery = getFirstInputText();
+        }
+
+        if (StringUtils.isBlank(searchQuery)) {
+            throw new IllegalArgumentException("未提供搜索关键词");
+        }
+        searchQuery = searchQuery.trim();
+        if (searchQuery.length() > 70) {
+            throw new IllegalArgumentException("搜索关键词不能超过 70 个字符");
+        }
+
+        log.info("Web search node processing, engine: {}, result_count: {}",
+            config.getSearchEngine(), config.getResultCount());
+
+        String nodeMessageTemplate = getNodeMessageTemplate(NodeMessageTemplateEnum.GOOGLE_SEARCH.getValue());
+        notifyAndStoreMessage(wfState, nodeMessageTemplate);
+
+        ZhipuWebSearchClient searchClient = SpringUtil.getBean(ZhipuWebSearchClient.class);
+        ZhipuWebSearchClient.SearchResponse response = searchClient.search(
+            searchQuery,
+            config,
+            UUID.randomUUID().toString()
+        );
+        String searchResult = JsonUtil.toJson(response);
+        if (searchResult == null) {
+            throw new IllegalStateException("搜索结果序列化失败");
+        }
+
+        log.info("Web search completed, result count: {}", response.count());
+        notifyAndStoreMessage(wfState, nodeMessageTemplate + "返回 " + response.count() + " 条结果");
+
+        List<NodeIOData> outputs = List.of(
+            NodeIOData.createByText(DEFAULT_OUTPUT_PARAM_NAME, "智谱网络搜索结果", searchResult)
+        );
+        return NodeProcessResult.builder().content(outputs).build();
+    }
+}
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/googleSearch/GoogleSearchNodeConfig.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/googleSearch/GoogleSearchNodeConfig.java
@@ -0,0 +1,44 @@
+package org.ruoyi.workflow.workflow.node.googleSearch;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import jakarta.validation.constraints.Max;
+import jakarta.validation.constraints.Min;
+import jakarta.validation.constraints.Pattern;
+import lombok.Data;
+
+@Data
+public class GoogleSearchNodeConfig {
+    /**
+     * 搜索查询关键词
+     */
+    private String query;
+
+    @JsonProperty("search_engine")
+    @Pattern(
+        regexp = "search_std|search_pro|search_pro_sogou|search_pro_quark",
+        message = "搜索引擎参数无效"
+    )
+    private String searchEngine = "search_std";
+
+    @JsonProperty("result_count")
+    @Min(value = 1, message = "搜索结果数量不能小于 1")
+    @Max(value = 50, message = "搜索结果数量不能大于 50")
+    private Integer resultCount = 10;
+
+    @JsonProperty("search_domain_filter")
+    private String searchDomainFilter;
+
+    @JsonProperty("search_recency_filter")
+    @Pattern(
+        regexp = "oneDay|oneWeek|oneMonth|oneYear|noLimit",
+        message = "搜索时间范围参数无效"
+    )
+    private String searchRecencyFilter = "noLimit";
+
+    @JsonProperty("content_size")
+    @Pattern(regexp = "medium|high", message = "网页摘要长度参数无效")
+    private String contentSize = "medium";
+
+    @JsonProperty("include_image")
+    private Boolean includeImage = false;
+}
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/googleSearch/ZhipuWebSearchClient.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/googleSearch/ZhipuWebSearchClient.java
@@ -0,0 +1,155 @@
+package org.ruoyi.workflow.workflow.node.googleSearch;
+
+import ai.z.openapi.ZhipuAiClient;
+import ai.z.openapi.service.web_search.WebSearchRequest;
+import ai.z.openapi.service.web_search.WebSearchResp;
+import ai.z.openapi.service.web_search.WebSearchResponse;
+import lombok.RequiredArgsConstructor;
+import org.apache.commons.lang3.StringUtils;
+import org.ruoyi.common.chat.domain.bo.chat.ChatModelBo;
+import org.ruoyi.common.chat.service.chat.IChatModelService;
+import org.springframework.stereotype.Component;
+
+import java.util.List;
+import java.util.concurrent.TimeUnit;
+
+/**
+ * 智谱 Web Search 官方 SDK 适配器。
+ */
+@Component
+@RequiredArgsConstructor
+public class ZhipuWebSearchClient {
+
+    private static final String ZHIPU_PROVIDER_CODE = "zhipu";
+    private static final String DEFAULT_BASE_URL = "https://open.bigmodel.cn/api/paas/v4/";
+
+    private final ZhipuWebSearchProperties properties;
+    private final IChatModelService chatModelService;
+
+    public SearchResponse search(String query, GoogleSearchNodeConfig config, String requestId) {
+        Credential credential = resolveCredential();
+        ZhipuAiClient client = createClient(credential);
+        try {
+            WebSearchRequest request = WebSearchRequest.builder()
+                .searchQuery(query)
+                .searchEngine(config.getSearchEngine())
+                .count(config.getResultCount())
+                .searchDomainFilter(blankToNull(config.getSearchDomainFilter()))
+                .searchRecencyFilter(config.getSearchRecencyFilter())
+                .contentSize(config.getContentSize())
+                .includeImage(config.getIncludeImage())
+                .requestId(requestId)
+                .build();
+
+            WebSearchResponse response = client.webSearch().createWebSearch(request);
+            if (response == null || !response.isSuccess() || response.getData() == null) {
+                String message = response == null ? "接口未返回响应" : StringUtils.defaultIfBlank(response.getMsg(), "未知错误");
+                throw new IllegalStateException("智谱 Web Search 调用失败：" + message);
+            }
+
+            List<SearchResult> results = response.getData().getWebSearchResp() == null
+                ? List.of()
+                : response.getData().getWebSearchResp().stream()
+                    .map(this::toSearchResult)
+                    .toList();
+
+            return new SearchResponse(
+                query,
+                config.getSearchEngine(),
+                response.getData().getRequestId(),
+                results.size(),
+                results
+            );
+        } finally {
+            client.close();
+        }
+    }
+
+    private ZhipuAiClient createClient(Credential credential) {
+        int connectTimeout = positiveOrDefault(properties.getConnectTimeout(), 10);
+        int readTimeout = positiveOrDefault(properties.getReadTimeout(), 30);
+        return ZhipuAiClient.builder()
+            .ofZHIPU()
+            .apiKey(credential.apiKey())
+            .baseUrl(credential.baseUrl())
+            .networkConfig(connectTimeout, readTimeout, readTimeout, readTimeout, TimeUnit.SECONDS)
+            .enableTokenCache()
+            .build();
+    }
+
+    private Credential resolveCredential() {
+        if (isUsableApiKey(properties.getApiKey())) {
+            return new Credential(normalizeBaseUrl(properties.getBaseUrl()), properties.getApiKey().trim());
+        }
+
+        ChatModelBo query = new ChatModelBo();
+        query.setProviderCode(ZHIPU_PROVIDER_CODE);
+        return chatModelService.queryList(query).stream()
+            .filter(model -> isUsableApiKey(model.getApiKey()))
+            .findFirst()
+            .map(model -> new Credential(normalizeBaseUrl(model.getApiHost()), model.getApiKey().trim()))
+            .orElseThrow(() -> new IllegalStateException(
+                "未配置智谱 API Key，请设置环境变量 ZAI_API_KEY，或在模型管理中配置 zhipu 厂商密钥"
+            ));
+    }
+
+    private boolean isUsableApiKey(String apiKey) {
+        return StringUtils.isNotBlank(apiKey)
+            && !"sk_xx".equalsIgnoreCase(apiKey.trim())
+            && !"your_api_key".equalsIgnoreCase(apiKey.trim());
+    }
+
+    private String normalizeBaseUrl(String baseUrl) {
+        String normalized = StringUtils.defaultIfBlank(baseUrl, DEFAULT_BASE_URL).trim();
+        normalized = StringUtils.removeEnd(normalized, "/");
+        if (!normalized.endsWith("/api/paas/v4")) {
+            normalized += "/api/paas/v4";
+        }
+        return normalized + "/";
+    }
+
+    private int positiveOrDefault(Integer value, int defaultValue) {
+        return value != null && value > 0 ? value : defaultValue;
+    }
+
+    private String blankToNull(String value) {
+        return StringUtils.isBlank(value) ? null : value.trim();
+    }
+
+    private SearchResult toSearchResult(WebSearchResp result) {
+        return new SearchResult(
+            result.getTitle(),
+            result.getContent(),
+            result.getLink(),
+            result.getMedia(),
+            result.getIcon(),
+            result.getRefer(),
+            result.getPublishDate(),
+            result.getImages()
+        );
+    }
+
+    private record Credential(String baseUrl, String apiKey) {
+    }
+
+    public record SearchResponse(
+        String query,
+        String searchEngine,
+        String requestId,
+        int count,
+        List<SearchResult> results
+    ) {
+    }
+
+    public record SearchResult(
+        String title,
+        String content,
+        String link,
+        String media,
+        String icon,
+        String refer,
+        String publishDate,
+        List<String> images
+    ) {
+    }
+}
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/googleSearch/ZhipuWebSearchProperties.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/googleSearch/ZhipuWebSearchProperties.java
@@ -0,0 +1,34 @@
+package org.ruoyi.workflow.workflow.node.googleSearch;
+
+import lombok.Data;
+import org.springframework.boot.context.properties.ConfigurationProperties;
+import org.springframework.stereotype.Component;
+
+/**
+ * 智谱 Web Search 配置。
+ */
+@Data
+@Component
+@ConfigurationProperties(prefix = "workflow.web-search.zhipu")
+public class ZhipuWebSearchProperties {
+
+    /**
+     * 智谱国内开放平台 API 根地址。
+     */
+    private String baseUrl = "https://open.bigmodel.cn/api/paas/v4/";
+
+    /**
+     * 智谱 API Key。建议通过环境变量 ZAI_API_KEY 注入。
+     */
+    private String apiKey;
+
+    /**
+     * 连接超时秒数。
+     */
+    private Integer connectTimeout = 10;
+
+    /**
+     * 读取超时秒数。
+     */
+    private Integer readTimeout = 30;
+}
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/humanFeedBack/HumanFeedbackNode.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/humanFeedBack/HumanFeedbackNode.java
@@ -1,56 +0,0 @@
-package org.ruoyi.workflow.workflow.node.humanFeedBack;
-
-import lombok.extern.slf4j.Slf4j;
-import org.apache.commons.lang3.StringUtils;
-import org.ruoyi.workflow.entity.WorkflowComponent;
-import org.ruoyi.workflow.entity.WorkflowNode;
-import org.ruoyi.workflow.workflow.NodeProcessResult;
-import org.ruoyi.workflow.workflow.WfNodeState;
-import org.ruoyi.workflow.workflow.WfState;
-import org.ruoyi.workflow.workflow.WorkflowUtil;
-import org.ruoyi.workflow.workflow.data.NodeIOData;
-import org.ruoyi.workflow.workflow.node.AbstractWfNode;
-
-import static org.ruoyi.workflow.cosntant.AdiConstant.WorkflowConstant.*;
-
-/**
- * 人机交互节点实现类
- */
-@Slf4j
-public class HumanFeedbackNode extends AbstractWfNode {
-
-    public HumanFeedbackNode(WorkflowComponent component, WorkflowNode nodeDefinition, WfState wfState, WfNodeState nodeState) {
-        super(component, nodeDefinition, wfState, nodeState);
-    }
-
-    // 人机交互节点的处理逻辑
-    @Override
-    public NodeProcessResult onProcess() {
-        log.info("Processing HumanFeedback node: {}", node.getTitle());
-        // 从状态中获取用户输入数据
-        Object humanFeedbackState = state.data().get(HUMAN_FEEDBACK_KEY);
-        if (null != humanFeedbackState) {
-            String userInput = humanFeedbackState.toString();
-            if (StringUtils.isNotBlank(userInput)) {
-                // 用户已提供输入，将用户输入添加到节点输入和输出中
-                NodeIOData feedbackData = NodeIOData.createByText("output", "default", userInput);
-                // 添加到输出列表，这样后续节点可以使用
-                state.getOutputs().add(feedbackData);
-                // 设置为成功状态
-                state.setProcessStatus(NODE_PROCESS_STATUS_SUCCESS);
-                log.info("Human feedback processed for node: {}, content: {}", node.getTitle(), userInput);
-            } else {
-                // 用户输入为空，设置等待状态
-                state.setProcessStatus(NODE_PROCESS_STATUS_DOING);
-                log.info("Human feedback is empty for node: {}", node.getTitle());
-            }
-        } else {
-            // 没有用户输入，这可能是正常情况（等待用户输入）
-            // 但为了确保流程可以继续，我们仍然标记为成功
-            state.setProcessStatus(NODE_PROCESS_STATUS_SUCCESS);
-            log.info("No human feedback found for node: {}, continuing workflow", node.getTitle());
-        }
-        return new NodeProcessResult();
-    }
-}
-
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/keywordExtractor/KeywordExtractorNode.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/keywordExtractor/KeywordExtractorNode.java
@@ -1,106 +0,0 @@
-package org.ruoyi.workflow.workflow.node.keywordExtractor;
-
-import dev.langchain4j.data.message.SystemMessage;
-import lombok.extern.slf4j.Slf4j;
-import org.apache.commons.lang3.StringUtils;
-import org.ruoyi.workflow.entity.WorkflowComponent;
-import org.ruoyi.workflow.entity.WorkflowNode;
-import org.ruoyi.workflow.util.SpringUtil;
-import org.ruoyi.workflow.util.WorkflowMessageUtil;
-import org.ruoyi.workflow.workflow.NodeProcessResult;
-import org.ruoyi.workflow.workflow.WfNodeState;
-import org.ruoyi.workflow.workflow.WfState;
-import org.ruoyi.workflow.workflow.WorkflowUtil;
-import org.ruoyi.workflow.workflow.data.NodeIOData;
-import org.ruoyi.workflow.workflow.node.AbstractWfNode;
-import org.ruoyi.workflow.workflow.node.enmus.NodeMessageTemplateEnum;
-
-import java.util.ArrayList;
-import java.util.List;
-
-import static org.ruoyi.workflow.cosntant.AdiConstant.WorkflowConstant.DEFAULT_OUTPUT_PARAM_NAME;
-
-/**
- * 【节点】关键词提取节点
- * 使用 LLM 从文本中提取关键词
- */
-@Slf4j
-public class KeywordExtractorNode extends AbstractWfNode {
-
-    public KeywordExtractorNode(WorkflowComponent wfComponent, WorkflowNode nodeDef, WfState wfState, WfNodeState nodeState) {
-        super(wfComponent, nodeDef, wfState, nodeState);
-    }
-
-    /**
-     * 处理关键词提取
-     * nodeConfig 格式：
-     * {
-     * "model_name": "deepseek-chat",
-     * "category": "llm",
-     * "top_n": 5,
-     * "prompt": "额外的提示词"
-     * }
-     *
-     * @return 提取的关键词列表
-     */
-    @Override
-    public NodeProcessResult onProcess() {
-        KeywordExtractorNodeConfig config = checkAndGetConfig(KeywordExtractorNodeConfig.class);
-
-        // 获取输入文本
-        String inputText = getFirstInputText();
-        if (StringUtils.isBlank(inputText)) {
-            log.warn("Keyword extractor node has no input text, node: {}", state.getUuid());
-            // 返回空结果
-            List<NodeIOData> outputs = new ArrayList<>();
-            outputs.add(NodeIOData.createByText(DEFAULT_OUTPUT_PARAM_NAME, "", ""));
-            return NodeProcessResult.builder().content(outputs).build();
-        }
-
-        log.info("Keyword extractor node config: {}", config);
-        log.info("Input text length: {}", inputText.length());
-
-        // 构建提示词
-        String prompt = buildPrompt(config, inputText);
-        log.info("Keyword extraction prompt: {}", prompt);
-
-        // 调用 LLM 进行关键词提取
-        WorkflowUtil workflowUtil = SpringUtil.getBean(WorkflowUtil.class);
-        String modelName = config.getModelName();
-        // 获取节点模板提示词信息
-        String nodeMessageTemplate = WorkflowMessageUtil.getNodeMessageTemplate(NodeMessageTemplateEnum.KEYWORD_EXTRACTOR.getValue());
-        // 发送SSE事件消息
-        WorkflowMessageUtil.sendEmitterMessage(wfState.getSseEmitter(), node, nodeMessageTemplate);
-        // 使用流式调用
-        workflowUtil.streamingInvokeLLM(wfState, state, node, modelName, prompt, nodeMessageTemplate);
-        return new NodeProcessResult();
-    }
-
-    /**
-     * 构建关键词提取的提示词
-     */
-    private String buildPrompt(KeywordExtractorNodeConfig config, String inputText) {
-        StringBuilder promptBuilder = new StringBuilder();
-
-        // 基础提示词
-        promptBuilder.append("请从以下文本中提取 ").append(config.getTopN()).append(" 个最重要的关键词。\n\n");
-
-        // 添加自定义提示词（如果有）
-        if (StringUtils.isNotBlank(config.getPrompt())) {
-            promptBuilder.append(config.getPrompt()).append("\n\n");
-        }
-
-        // 输出格式要求
-        promptBuilder.append("要求：\n");
-        promptBuilder.append("1. 只返回关键词，每个关键词用逗号分隔\n");
-        promptBuilder.append("2. 关键词应该是名词或名词短语\n");
-        promptBuilder.append("3. 按重要性从高到低排序\n");
-        promptBuilder.append("4. 不要添加任何解释或额外的文字\n\n");
-
-        // 原始文本
-        promptBuilder.append("文本内容：\n");
-        promptBuilder.append(inputText);
-
-        return promptBuilder.toString();
-    }
-}
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/keywordExtractor/KeywordExtractorNodeConfig.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/keywordExtractor/KeywordExtractorNodeConfig.java
@@ -1,42 +0,0 @@
-package org.ruoyi.workflow.workflow.node.keywordExtractor;
-
-import com.fasterxml.jackson.annotation.JsonProperty;
-import jakarta.validation.constraints.Max;
-import jakarta.validation.constraints.Min;
-import jakarta.validation.constraints.NotNull;
-import lombok.Data;
-import lombok.EqualsAndHashCode;
-
-/**
- * 关键词提取节点配置
- */
-@EqualsAndHashCode
-@Data
-public class KeywordExtractorNodeConfig {
-
-    /**
-     * 模型分类（如：llm, embedding 等）
-     */
-    private String category;
-
-    /**
-     * 模型名称
-     */
-    @NotNull
-    @JsonProperty("model_name")
-    private String modelName;
-
-    /**
-     * 提取的关键词数量
-     */
-    @Min(1)
-    @Max(50)
-    @JsonProperty("top_n")
-    private Integer topN = 5;
-
-    /**
-     * 提示词（可选）
-     * 用于指导关键词提取的额外说明
-     */
-    private String prompt;
-}
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/mailSend/MailSendNode.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/mailSend/MailSendNode.java
@@ -7,6 +7,7 @@ import lombok.extern.slf4j.Slf4j;
 import org.apache.commons.lang3.StringUtils;
 import org.ruoyi.workflow.entity.WorkflowComponent;
 import org.ruoyi.workflow.entity.WorkflowNode;
+import org.ruoyi.workflow.util.JsonUtil;
 import org.ruoyi.workflow.workflow.NodeProcessResult;
 import org.ruoyi.workflow.workflow.WfNodeState;
 import org.ruoyi.workflow.workflow.WfState;
@@ -40,15 +41,25 @@ public class MailSendNode extends AbstractWfNode {
            String input = getDataFromInput(inputs);
            // 判断是否为JSON格式(LLM输出转换 由LLM生成格式)
            if (StringUtils.isNotBlank(input) && isJson(input)) {
-                // 使用Jackson解析和合并配置
-                ObjectMapper objectMapper = new ObjectMapper();
-                JsonNode inputJson = objectMapper.readTree(input);
-                // 将config转换为JsonNode
-                JsonNode configJson = objectMapper.valueToTree(config);
-                // 合并两个JSON节点
-                JsonNode mergedJson = objectMapper.readerForUpdating(configJson).readValue(inputJson);
-                // 转换回config对象
-                config = objectMapper.treeToValue(mergedJson, MailSendNodeConfig.class);
+                try {
+                    // 使用统一的 JsonUtil 进行解析
+                    JsonNode inputJson = JsonUtil.toJsonNode(input);
+                    if (inputJson != null) {
+                        // 使用 JsonUtil 内部的 ObjectMapper 进行合并
+                        ObjectMapper objectMapper = new ObjectMapper();
+                        // 将config转换为JsonNode
+                        JsonNode configJson = objectMapper.valueToTree(config);
+                        // 合并两个JSON节点
+                        JsonNode mergedJson = objectMapper.readerForUpdating(configJson).readValue(inputJson);
+                        // 转换回config对象
+                        config = objectMapper.treeToValue(mergedJson, MailSendNodeConfig.class);
+                    } else {
+                        log.warn("输入 JSON 解析结果为 null，使用原始配置");
+                    }
+                } catch (Exception e) {
+                    log.error("合并邮件配置失败，使用原始配置: {}", e.getMessage(), e);
+                    // 继续使用原始 config，不中断流程
+                }
            }

            // 安全获取模板（使用 defaultString 避免 null）
--- a/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/switcher/SwitcherNode.java
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/java/org/ruoyi/workflow/workflow/node/switcher/SwitcherNode.java
@@ -1,7 +1,6 @@
 package org.ruoyi.workflow.workflow.node.switcher;

 import com.fasterxml.jackson.databind.JsonNode;
-import com.fasterxml.jackson.databind.ObjectMapper;
 import lombok.extern.slf4j.Slf4j;
 import org.apache.commons.lang3.ObjectUtils;
 import org.apache.commons.lang3.StringUtils;
@@ -9,6 +8,7 @@ import org.ruoyi.common.core.utils.SpringUtils;
 import org.ruoyi.workflow.entity.WorkflowComponent;
 import org.ruoyi.workflow.entity.WorkflowNode;
 import org.ruoyi.workflow.service.WorkflowNodeService;
+import org.ruoyi.workflow.util.JsonUtil;
 import org.ruoyi.workflow.workflow.NodeProcessResult;
 import org.ruoyi.workflow.workflow.WfNodeState;
 import org.ruoyi.workflow.workflow.WfState;
@@ -18,8 +18,6 @@ import org.ruoyi.workflow.workflow.node.enmus.NodeMessageTemplateEnum;

 import java.math.BigDecimal;
 import java.util.List;
-import java.util.Objects;
-import java.util.Optional;

 /**
 * 条件分支节点
@@ -339,9 +337,12 @@ public class SwitcherNode extends AbstractWfNode {
            log.info("节点 '{}' 的输入配置: {}", nodeUuid, inputConfig);
            if (StringUtils.isNotBlank(inputConfig)){
                try {
-                    // 使用Jackson解析输入配置
-                    ObjectMapper objectMapper = new ObjectMapper();
-                    JsonNode configJson = objectMapper.readTree(inputConfig);
+                    // 使用统一的 JsonUtil 而不是每次创建新的 ObjectMapper
+                    JsonNode configJson = JsonUtil.toJsonNode(inputConfig);
+                    if (configJson == null) {
+                        log.warn("节点 '{}' 的输入配置 JSON 解析结果为 null", nodeUuid);
+                        return result;
+                    }
                    // 获取 user_inputs 数组
                    JsonNode userInputs = configJson.get("user_inputs");
                    if (userInputs != null && userInputs.isArray()) {
@@ -357,7 +358,8 @@ public class SwitcherNode extends AbstractWfNode {
                        }
                    }
                } catch (Exception e) {
-                    log.error("解析节点输入配置失败: {}", nodeUuid, e);
+                    log.error("解析节点 '{}' 输入配置失败，参数名: {}, 配置内容: {}", nodeUuid, paramName, inputConfig, e);
+                    // 不抛出异常，返回默认结果，避免中断整个流程
                }
            }
        }
--- a/ruoyi-modules/ruoyi-aiflow/src/main/resources/sql/workflow_nodes_config.sql
+++ b/ruoyi-modules/ruoyi-aiflow/src/main/resources/sql/workflow_nodes_config.sql
@@ -0,0 +1,21 @@
+-- =============================================
+-- 流程编排搜索节点配置脚本
+-- =============================================
+-- 说明：本脚本用于添加搜索节点的系统配置
+-- 执行前请确保 sys_config 表存在
+-- =============================================
+
+-- 搜索节点模板配置
+INSERT INTO sys_config (config_name, config_key, config_value, config_type, remark, create_by, create_time, update_by, update_time)
+VALUES ('搜索节点模板', 'node.googleSearch.template', '正在搜索相关内容...', 'Y', '搜索节点的响应模板，用于网络搜索功能', 'admin', NOW(), 'admin', NOW())
+ON DUPLICATE KEY UPDATE config_value = '正在搜索相关内容...', update_time = NOW();
+
+-- =============================================
+-- 验证配置是否添加成功
+-- =============================================
+SELECT config_id, config_name, config_key, config_value, config_type, remark
+FROM sys_config
+WHERE config_key IN (
+    'node.googleSearch.template'
+)
+ORDER BY config_id;
--- a/ruoyi-modules/ruoyi-aiflow/流程编排模块说明.md
+++ b/ruoyi-modules/ruoyi-aiflow/流程编排模块说明.md
@@ -1,425 +0,0 @@
-# Ruoyi-AI 流程编排模块详细说明文档
-
-## 概述
-
-Ruoyi-AI 工作流模块是一个基于 LangGraph4j 的智能工作流引擎，支持可视化工作流设计、AI 模型集成、条件分支、人机交互等高级功能。该模块采用微服务架构，提供完整的
-RESTful API 和流式响应支持。
-
-## 模块架构
-
-### 1. 核心依赖
-
- **LangGraph4j**: 1.5.3 - 工作流图执行引擎
- **LangChain4j**: 1.11.0 - AI 模型集成框架
- **Spring Boot**: 3.5.8 - 应用框架
- **MyBatis Plus**: 数据访问层
- **Redis**: 缓存和状态管理
- **OpenAPI**: API 文档
-
-## 核心功能
-
-### 1. 工作流管理
-
-#### 1.1 工作流定义
-
- **创建工作流**: 支持自定义标题、描述、公开性设置
- **编辑工作流**: 可视化节点编辑、连接线配置
- **版本控制**: 支持工作流的版本管理和回滚
- **权限管理**: 支持公开/私有工作流设置
-
-#### 1.2 工作流执行
-
- **流式执行**: 基于 SSE 的实时流式响应
- **状态管理**: 完整的执行状态跟踪
- **错误处理**: 详细的错误信息和异常处理
- **中断恢复**: 支持工作流中断和恢复执行
-
-### 2. 节点类型
-
-#### 2.1 基础节点
-
- **Start**: 开始节点，定义工作流入口
- **End**: 结束节点，定义工作流出口
-
-
-#### 2.2 AI 模型节点
-
- **Answer**: 大语言模型问答节点
- **Dalle3**: DALL-E 3 图像生成
- **Tongyiwanx**: 通义万相图像生成
- **Classifier**: 内容分类节点
-
-#### 2.3 数据处理节点
-
- **DocumentExtractor**: 文档信息提取
- **KeywordExtractor**: 关键词提取
- **FaqExtractor**: 常见问题提取
- **KnowledgeRetrieval**: 知识库检索
-
-#### 2.4 控制流节点
-
- **Switcher**: 条件分支节点
- **HumanFeedback**: 人机交互节点
-
-#### 2.5 外部集成节点
-
- **Google**: Google 搜索集成
- **MailSend**: 邮件发送
- **HttpRequest**: HTTP 请求
- **Template**: 模板转换
-
-### 3. 数据流管理
-
-#### 3.1 输入输出定义
-
-```java
-// 节点输入输出数据结构
-public class NodeIOData {
-    private String name;           // 参数名称
-    private NodeIODataContent content; // 参数内容
-}
-
-// 支持的数据类型
-public enum WfIODataTypeEnum {
-    TEXT,        // 文本
-    NUMBER,      // 数字
-    BOOLEAN,     // 布尔值
-    FILES,       // 文件
-    OPTIONS      // 选项
-}
-```
-
-#### 3.2 参数引用
-
- **节点间引用**: 支持上游节点输出作为下游节点输入
- **参数映射**: 自动处理参数名称映射
- **类型转换**: 自动进行数据类型转换
-
-## 数据库设计
-
-### 1. 核心表结构
-
-#### 1.1 工作流定义表 (t_workflow)
-
-```sql
-CREATE TABLE t_workflow (
-    id          BIGINT AUTO_INCREMENT PRIMARY KEY,
-    uuid        VARCHAR(32)  NOT NULL DEFAULT '',
-    title       VARCHAR(100) NOT NULL DEFAULT '',
-    remark      TEXT         NOT NULL DEFAULT '',
-    user_id     BIGINT       NOT NULL DEFAULT 0,
-    is_public   TINYINT(1)   NOT NULL DEFAULT 0,
-    is_enable   TINYINT(1)   NOT NULL DEFAULT 1,
-    create_time DATETIME     NOT NULL DEFAULT CURRENT_TIMESTAMP,
-    update_time DATETIME     NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
-    is_deleted  TINYINT(1)   NOT NULL DEFAULT 0
-);
-```
-
-#### 1.2 工作流节点表 (t_workflow_node)
-
-```sql
-CREATE TABLE t_workflow_node (
-    id                    BIGINT AUTO_INCREMENT PRIMARY KEY,
-    uuid                  VARCHAR(32)  NOT NULL DEFAULT '',
-    workflow_id           BIGINT       NOT NULL DEFAULT 0,
-    workflow_component_id BIGINT       NOT NULL DEFAULT 0,
-    user_id               BIGINT       NOT NULL DEFAULT 0,
-    title                 VARCHAR(100) NOT NULL DEFAULT '',
-    remark                VARCHAR(500) NOT NULL DEFAULT '',
-    input_config          JSON         NOT NULL DEFAULT ('{}'),
-    node_config           JSON         NOT NULL DEFAULT ('{}'),
-    position_x            DOUBLE       NOT NULL DEFAULT 0,
-    position_y            DOUBLE       NOT NULL DEFAULT 0,
-    create_time           DATETIME     NOT NULL DEFAULT CURRENT_TIMESTAMP,
-    update_time           DATETIME     NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
-    is_deleted            TINYINT(1)   NOT NULL DEFAULT 0
-);
-```
-
-#### 1.3 工作流边表 (t_workflow_edge)
-
-```sql
-CREATE TABLE t_workflow_edge (
-    id               BIGINT AUTO_INCREMENT PRIMARY KEY,
-    uuid             VARCHAR(32) NOT NULL DEFAULT '',
-    workflow_id      BIGINT      NOT NULL DEFAULT 0,
-    source_node_uuid VARCHAR(32) NOT NULL DEFAULT '',
-    source_handle    VARCHAR(32) NOT NULL DEFAULT '',
-    target_node_uuid VARCHAR(32) NOT NULL DEFAULT '',
-    create_time      DATETIME    NOT NULL DEFAULT CURRENT_TIMESTAMP,
-    update_time      DATETIME    NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
-    is_deleted       TINYINT(1)  NOT NULL DEFAULT 0
-);
-```
-
-#### 1.4 工作流运行时表 (t_workflow_runtime)
-
-```sql
-CREATE TABLE t_workflow_runtime (
-    id            BIGINT AUTO_INCREMENT PRIMARY KEY,
-    uuid          VARCHAR(32)  NOT NULL DEFAULT '',
-    user_id       BIGINT       NOT NULL DEFAULT 0,
-    workflow_id   BIGINT       NOT NULL DEFAULT 0,
-    input         JSON         NOT NULL DEFAULT ('{}'),
-    output        JSON         NOT NULL DEFAULT ('{}'),
-    status        SMALLINT     NOT NULL DEFAULT 1,
-    status_remark VARCHAR(250) NOT NULL DEFAULT '',
-    create_time   DATETIME     NOT NULL DEFAULT CURRENT_TIMESTAMP,
-    update_time   DATETIME     NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
-    is_deleted    TINYINT(1)   NOT NULL DEFAULT 0
-);
-```
-
-#### 1.5 工作流组件表 (t_workflow_component)
-
-```sql
-CREATE TABLE t_workflow_component (
-    id            BIGINT AUTO_INCREMENT PRIMARY KEY,
-    uuid          VARCHAR(32)  DEFAULT '' NOT NULL,
-    name          VARCHAR(32)  DEFAULT '' NOT NULL,
-    title         VARCHAR(100) DEFAULT '' NOT NULL,
-    remark        TEXT                   NOT NULL,
-    display_order INT          DEFAULT 0 NOT NULL,
-    is_enable     TINYINT(1)   DEFAULT 0 NOT NULL,
-    create_time   DATETIME     DEFAULT CURRENT_TIMESTAMP NOT NULL,
-    update_time   DATETIME     DEFAULT CURRENT_TIMESTAMP NOT NULL,
-    is_deleted    TINYINT(1)   DEFAULT 0 NOT NULL
-);
-```
-
-## API 接口
-
-### 1. 工作流管理接口
-
-#### 1.1 基础操作
-
-```http
-# 创建工作流
-POST /workflow/add
-Content-Type: application/json
-{
-    "title": "工作流标题",
-    "remark": "工作流描述",
-    "isPublic": false
-}
-
-# 更新工作流
-POST /workflow/update
-Content-Type: application/json
-{
-    "uuid": "工作流UUID",
-    "title": "新标题",
-    "remark": "新描述"
-}
-
-# 删除工作流
-POST /workflow/del/{uuid}
-
-# 启用/禁用工作流
-POST /workflow/enable/{uuid}?enable=true
-```
-
-#### 1.2 搜索和查询
-
-```http
-# 搜索我的工作流
-GET /workflow/mine/search?keyword=关键词&isPublic=true&currentPage=1&pageSize=10
-
-# 搜索公开工作流
-GET /workflow/public/search?keyword=关键词&currentPage=1&pageSize=10
-
-# 获取工作流组件列表
-GET /workflow/public/component/list
-```
-
-### 2. 工作流执行接口
-
-#### 2.1 流式执行
-
-```http
-# 流式执行工作流
-POST /workflow/run
-Content-Type: application/json
-Accept: text/event-stream
-{
-    "uuid": "工作流UUID",
-    "inputs": [
-        {
-            "name": "input",
-            "content": {
-                "type": 1,
-                "textContent": "用户输入内容"
-            }
-        }
-    ]
-}
-```
-
-#### 2.2 运行时管理
-
-```http
-# 恢复中断的工作流
-POST /workflow/runtime/resume/{runtimeUuid}
-Content-Type: application/json
-{
-    "feedbackContent": "用户反馈内容"
-}
-
-# 查询工作流执行历史
-GET /workflow/runtime/page?wfUuid=工作流UUID&currentPage=1&pageSize=10
-
-# 查询运行时节点详情
-GET /workflow/runtime/nodes/{runtimeUuid}
-
-# 清理运行时数据
-POST /workflow/runtime/clear?wfUuid=工作流UUID
-```
-
-### 3. 管理端接口
-
-#### 3.1 工作流管理
-
-```http
-# 搜索所有工作流
-POST /admin/workflow/search
-Content-Type: application/json
-{
-    "title": "搜索关键词",
-    "isPublic": true,
-    "isEnable": true
-}
-
-# 启用/禁用工作流
-POST /admin/workflow/enable?uuid=工作流UUID&isEnable=true
-```
-
-## 核心实现
-
-### 1. 工作流引擎 (WorkflowEngine)
-
-工作流引擎是整个模块的核心，负责：
-
- 工作流图的构建和编译
- 节点执行调度
- 状态管理和持久化
- 流式输出处理
-
-```java
-public class WorkflowEngine {
-    // 核心执行方法
-    public void run(User user, List<ObjectNode> userInputs, SseEmitter sseEmitter) {
-        // 1. 验证工作流状态
-        // 2. 创建运行时实例
-        // 3. 构建状态图
-        // 4. 执行工作流
-        // 5. 处理流式输出
-    }
-    
-    // 恢复执行方法
-    public void resume(String userInput) {
-        // 1. 更新状态
-        // 2. 继续执行
-    }
-}
-```
-
-### 2. 节点工厂 (WfNodeFactory)
-
-节点工厂负责根据组件类型创建对应的节点实例：
-
-```java
-public class WfNodeFactory {
-    public static AbstractWfNode create(WorkflowComponent component, 
-                                      WorkflowNode node, 
-                                      WfState wfState, 
-                                      WfNodeState nodeState) {
-        // 根据组件类型创建对应的节点实例
-        switch (component.getName()) {
-            case "Answer":
-                return new LLMAnswerNode(component, node, wfState, nodeState);
-            case "Switcher":
-                return new SwitcherNode(component, node, wfState, nodeState);
-            // ... 其他节点类型
-        }
-    }
-}
-```
-
-### 3. 图构建器 (WorkflowGraphBuilder)
-
-图构建器负责将工作流定义转换为可执行的状态图：
-
-```java
-public class WorkflowGraphBuilder {
-    public StateGraph<WfNodeState> build(WorkflowNode startNode) {
-        // 1. 构建编译节点树
-        // 2. 转换为状态图
-        // 3. 添加节点和边
-        // 4. 处理条件分支
-        // 5. 处理并行执行
-    }
-}
-```
-
-## 流式响应机制
-
-### 1. SSE 事件类型
-
-工作流执行过程中会发送多种类型的 SSE 事件：
-
-```javascript
-// 节点开始执行
-[NODE_RUN_节点UUID] - 节点执行开始事件
-
-// 节点输入数据
-[NODE_INPUT_节点UUID] - 节点输入数据事件
-
-// 节点输出数据
-[NODE_OUTPUT_节点UUID] - 节点输出数据事件
-
-// 流式内容块
-[NODE_CHUNK_节点UUID] - 流式内容块事件
-
-// 等待用户输入
-[NODE_WAIT_FEEDBACK_BY_节点UUID] - 等待用户输入事件
-```
-
-### 2. 流式处理流程
-
-1. **初始化**: 创建工作流运行时实例
-2. **节点执行**: 逐个执行工作流节点
-3. **实时输出**: 通过 SSE 实时推送执行结果
-4. **状态更新**: 实时更新节点和工作流状态
-5. **错误处理**: 捕获并处理执行过程中的错误
-
-## 扩展开发
-
-### 1. 自定义节点开发
-
-要开发自定义工作流节点，需要：
-
-1. **创建节点类**：继承 `AbstractWfNode`
-2. **实现处理逻辑**：重写 `onProcess()` 方法
-3. **定义配置类**：创建节点配置类
-4. **注册组件**：在组件表中注册新组件
-
-```java
-public class CustomNode extends AbstractWfNode {
-    @Override
-    protected NodeProcessResult onProcess() {
-        // 实现自定义处理逻辑
-        List<NodeIOData> outputs = new ArrayList<>();
-        // ... 处理逻辑
-        return NodeProcessResult.success(outputs);
-    }
-}
-```
-
-### 2. 自定义组件注册
-
-```sql
-- 在 t_workflow_component 表中添加新组件
-INSERT INTO t_workflow_component (uuid, name, title, remark, is_enable) 
-VALUES (REPLACE(UUID(), '-', ''), 'CustomNode', '自定义节点', '自定义节点描述', true);
-```
--- a/ruoyi-modules/ruoyi-chat/src/main/java/org/ruoyi/controller/chat/ChatController.java
+++ b/ruoyi-modules/ruoyi-chat/src/main/java/org/ruoyi/controller/chat/ChatController.java
@@ -1,10 +1,8 @@
 package org.ruoyi.controller.chat;

-import jakarta.servlet.http.HttpServletRequest;
 import jakarta.validation.Valid;
 import lombok.RequiredArgsConstructor;
 import lombok.extern.slf4j.Slf4j;
-import org.ruoyi.common.chat.domain.dto.request.AgentChatRequest;
 import org.ruoyi.common.chat.domain.dto.request.ChatRequest;
 import org.ruoyi.service.chat.impl.ChatServiceFacade;
 import org.springframework.stereotype.Controller;
--- a/ruoyi-modules/ruoyi-chat/src/main/java/org/ruoyi/service/chat/impl/ChatServiceFacade.java
+++ b/ruoyi-modules/ruoyi-chat/src/main/java/org/ruoyi/service/chat/impl/ChatServiceFacade.java
@@ -150,35 +150,59 @@ public class ChatServiceFacade implements IChatService {
     * @return SseEmitter
     */
    public SseEmitter sseChat(ChatRequest chatRequest) {
-
-        // 具体的服务实现
        Long userId = LoginHelper.getUserId();
        String tokenValue = StpUtil.getTokenValue();
-        // 每个会话一个 SSE 连接，避免同用户多会话串台
-        SseEmitter emitter = sseEmitterManager.connect(String.valueOf(chatRequest.getSessionId()));
+
+        boolean workflowMode = Boolean.TRUE.equals(chatRequest.getEnableWorkFlow());
+        boolean agentMode = chatRequest.getAgentId() != null;
+        if (workflowMode && agentMode) {
+            throw new IllegalArgumentException("对话模式参数冲突：工作流和智能体不能同时启用");
+        }
+
+        // 工作流模式。工作流引擎负责创建并持有自己的 SSE，必须在普通聊天连接创建前路由。
+        if (workflowMode) {
+            chatMessageService.saveChatMessage(
+                userId,
+                chatRequest.getSessionId(),
+                chatRequest.getContent(),
+                RoleType.USER.getName(),
+                chatRequest.getModel()
+            );
+            return handleWorkflowChat(chatRequest);
+        }

        // 智能体解析：传入 agentId 时按智能体绑定的模型覆盖 model 字段
        AgentVo agentVo = null;
-        if (chatRequest.getAgentId() != null) {
+        if (agentMode) {
            agentVo = agentService.queryById(chatRequest.getAgentId());
+            if (agentVo == null) {
+                throw new IllegalArgumentException("智能体不存在: " + chatRequest.getAgentId());
+            }
            if (agentVo != null && agentVo.getModelId() != null) {
                ChatModelVo agentModel = chatModelService.queryById(agentVo.getModelId());
-                if (agentModel != null) {
-                    chatRequest.setModel(agentModel.getModelName());
+                if (agentModel == null) {
+                    throw new IllegalArgumentException("智能体绑定的模型不存在: " + agentVo.getModelId());
                }
-            } else {
-                log.warn("智能体不存在或未配置模型，回退到 model 字段: agentId={}", chatRequest.getAgentId());
+                chatRequest.setModel(agentModel.getModelName());
            }
        }

+        if (StringUtils.isBlank(chatRequest.getModel())) {
+            throw new IllegalArgumentException(
+                agentVo == null ? "对话模式必须指定模型" : "智能体未绑定模型，且请求未提供回退模型"
+            );
+        }
+
        // 根据模型名称查询完整配置
        ChatModelVo chatModelVo = chatModelService.selectModelByName(chatRequest.getModel());
        if (chatModelVo == null) {
            throw new IllegalArgumentException("模型不存在: " + chatRequest.getModel());
        }

+        // 对话和智能体模式共用按会话隔离的 SSE。
+        SseEmitter emitter = sseEmitterManager.connect(String.valueOf(chatRequest.getSessionId()));
+
        // 构建上下文消息列表（系统提示词 + 历史消息 + 当前用户消息）
-        // 注意：RAG 检索增强统一在 handleAgentChat 中执行一次，此处不再重复检索
        List<ChatMessage> contextMessages = buildContextMessages(chatRequest, agentVo);

        chatRequest.setEmitter(emitter);
@@ -190,43 +214,68 @@ public class ChatServiceFacade implements IChatService {
        // 保存用户消息
        chatMessageService.saveChatMessage(userId, chatRequest.getSessionId(), chatRequest.getContent(), RoleType.USER.getName(), chatRequest.getModel());

-        TraceRunHandle traceRun = Boolean.TRUE.equals(chatRequest.getEnableWorkFlow())
-            ? null : startRagTraceRun(chatRequest, userId);
-        // 3. 路由对话模式：工作流对话 / 智能体对话（两者均返回各自的 SseEmitter）
-        return handleSpecialChatModes(chatRequest, agentVo, traceRun);
-    }
+        TraceRunHandle traceRun = startRagTraceRun(chatRequest, userId);

-    /**
-     * 路由对话模式：仅两种情况——工作流对话 / 智能体对话。
-     *
-     * @param chatRequest 聊天请求
-     * @param agentVo    智能体配置（可为 null）
-     * @return 对应模式的 SseEmitter
-     */
-    private SseEmitter handleSpecialChatModes(ChatRequest chatRequest, AgentVo agentVo, TraceRunHandle traceRun) {
-        // 模式1：工作流对话（前端应用市场选工作流后携带 workFlowRunner）
-        if (Boolean.TRUE.equals(chatRequest.getEnableWorkFlow())) {
-            log.info("处理工作流对话,会话: {}", chatRequest.getSessionId());
-            WorkFlowRunner runner = chatRequest.getWorkFlowRunner();
-            if (ObjectUtils.isEmpty(runner)) {
-                log.warn("工作流参数为空");
-            }
-            return workFlowStarterService.streaming(
-                ThreadContext.getCurrentUser(),
-                runner.getUuid(),
-                runner.getInputs(),
-                chatRequest.getSessionId()
-            );
+        // 智能体和普通对话互斥：有 agentId 为智能体，否则为普通模型对话。
+        if (agentVo != null) {
+            log.info("处理智能体对话,会话:{},agentId:{}", chatRequest.getSessionId(), chatRequest.getAgentId());
+            return handleAgentChat(chatRequest, agentVo, traceRun);
        }
-        // 模式2：智能体对话（默认走 Supervisor 多 Agent 编排）
-        return handleAgentChat(chatRequest, agentVo, traceRun);
+        log.info("处理普通对话,会话:{},模型:{}", chatRequest.getSessionId(), chatRequest.getModel());
+        return handleModelChat(chatRequest, traceRun);
    }

    /**
-     * 智能体对话模式（默认）：构建 Supervisor 多 Agent 编排并异步执行，结果通过 SSE 推送。
+     * 工作流模式。工作流运行时负责 SSE、节点执行和结束事件。
+     */
+    private SseEmitter handleWorkflowChat(ChatRequest chatRequest) {
+        WorkFlowRunner runner = chatRequest.getWorkFlowRunner();
+        if (ObjectUtils.isEmpty(runner) || StringUtils.isBlank(runner.getUuid())) {
+            throw new IllegalArgumentException("工作流模式必须提供 workFlowRunner.uuid");
+        }
+        log.info("处理工作流对话,会话:{},workflowUuid:{}", chatRequest.getSessionId(), runner.getUuid());
+        return workFlowStarterService.streaming(
+            ThreadContext.getCurrentUser(),
+            runner.getUuid(),
+            runner.getInputs() == null ? List.of() : runner.getInputs(),
+            chatRequest.getSessionId()
+        );
+    }
+
+    /**
+     * 普通对话模式：直接调用选定模型，不装配 Supervisor、MCP、Skills 或专业子 Agent。
+     */
+    private SseEmitter handleModelChat(ChatRequest chatRequest, TraceRunHandle traceRun) {
+        ChatModelVo chatModelVo = chatRequest.getChatModelVo();
+        AbstractChatService chatService = chatServiceFactory.getOriginalService(chatModelVo.getProviderCode());
+        StreamingChatModel streamingChatModel = chatService.buildStreamingChatModel(chatModelVo, chatRequest);
+        List<ChatMessage> messages = buildModelChatMessages(chatRequest);
+
+        TraceStreamSpan llmSpan = null;
+        try (TraceScope ignored = openTraceScope(traceRun, chatRequest.getUserId())) {
+            llmSpan = startLlmCallSpan(traceRun, chatRequest, "handleModelChat");
+            streamingChatModel.chat(
+                messages,
+                createModelChatResponseHandler(chatRequest, traceRun, llmSpan)
+            );
+        } catch (Exception e) {
+            if (llmSpan != null) {
+                llmSpan.finishError(e);
+                llmSpan.detach();
+            }
+            finishTraceRun(traceRun, TraceConstants.STATUS_ERROR, e);
+            SseMessageUtils.sendError(String.valueOf(chatRequest.getSessionId()), e.getMessage());
+            SseMessageUtils.completeConnection(String.valueOf(chatRequest.getSessionId()));
+            log.error("普通对话执行失败", e);
+        }
+        return chatRequest.getEmitter();
+    }
+
+    /**
+     * 智能体对话模式：构建 Supervisor 多 Agent 编排并异步执行，结果通过 SSE 推送。
     *
     * @param chatRequest 聊天请求
-     * @param agentVo    智能体配置（可为 null，无智能体时用请求 model 兜底）
+     * @param agentVo    智能体配置
     */
    private SseEmitter handleAgentChat(ChatRequest chatRequest, AgentVo agentVo, TraceRunHandle traceRun) {
        ChatModelVo chatModelVo = chatRequest.getChatModelVo();
@@ -321,7 +370,7 @@ public class ChatServiceFacade implements IChatService {
        CompletableFuture.runAsync(() -> {
            TraceStreamSpan llmSpan = null;
            try (TraceScope ignored = openTraceScope(traceRun, userId)) {
-                llmSpan = startLlmCallSpan(traceRun, chatRequest);
+                llmSpan = startLlmCallSpan(traceRun, chatRequest, "handleAgentChat");
                String result = supervisor.invoke(prompt);
                SseMessageUtils.sendContent(sessionId, result);
                SseMessageUtils.sendDone(sessionId);
@@ -386,7 +435,8 @@ public class ChatServiceFacade implements IChatService {
            traceRun.businessId, userId, traceRun.tenantId);
    }

-    private TraceStreamSpan startLlmCallSpan(TraceRunHandle traceRun, ChatRequest chatRequest) {
+    private TraceStreamSpan startLlmCallSpan(TraceRunHandle traceRun, ChatRequest chatRequest,
+                                             String methodName) {
        if (traceRun == null || StringUtils.isBlank(TraceContext.getTraceId())) {
            return null;
        }
@@ -401,7 +451,7 @@ public class ChatServiceFacade implements IChatService {
        node.setNodeName("llm-call");
        node.setNodeType(RagTraceNodeTypes.NODE_LLM_CALL);
        node.setClassName(ChatServiceFacade.class.getName());
-        node.setMethodName("handleAgentChat");
+        node.setMethodName(methodName);
        node.setStatus(TraceConstants.STATUS_RUNNING);
        node.setStartTime(new Date(startMillis));
        node.setInputPayload(RagTracePayloadBuilder.streamInputSummary(chatRequest));
@@ -564,7 +614,11 @@ public class ChatServiceFacade implements IChatService {
        Long userId = LoginHelper.getUserId();

        // 5. 建立 SSE 连接（用于前端监听，按会话隔离）
-        sseEmitterManager.connect(String.valueOf(chatRequest.getSessionId()));
+        // 工作流调用时(externalHandler 非空), SSE 连接由工作流引擎创建并持有(WorkflowStarter#streaming),
+        // connect 为替换语义(关闭同键旧连接), 此处重连会掐断工作流连接, 必须跳过
+        if (externalHandler == null) {
+            sseEmitterManager.connect(String.valueOf(chatRequest.getSessionId()));
+        }

        // 保存用户消息
        chatMessageService.saveChatMessage(userId, chatRequest.getSessionId(), chatRequest.getContent(), RoleType.USER.getName(), chatRequest.getModel());
@@ -645,6 +699,19 @@ public class ChatServiceFacade implements IChatService {
        return messages;
    }

+    /**
+     * 构建普通对话消息。保留历史上下文，并在请求指定知识库时仅增强当前用户消息。
+     */
+    private List<ChatMessage> buildModelChatMessages(ChatRequest chatRequest) {
+        List<ChatMessage> messages = new ArrayList<>(chatRequest.getContextMessages());
+        String augmentedInput = augmentAgentInput(chatRequest, null);
+        int lastIndex = messages.size() - 1;
+        if (lastIndex >= 0 && messages.get(lastIndex) instanceof UserMessage) {
+            messages.set(lastIndex, UserMessage.userMessage(augmentedInput));
+        }
+        return messages;
+    }
+
    /**
     * 将上下文消息格式化为多轮对话文本（供只接受 String 输入的 Supervisor 使用）。
     * 跳过 SystemMessage（系统提示词单独前置）与最后一条当前用户消息（单独做 RAG 增强后拼接）。
@@ -791,6 +858,77 @@ public class ChatServiceFacade implements IChatService {
        return queryVectorBo;
    }

+    /**
+     * 普通对话响应处理器：推送流式内容、保存助手消息并结束链路追踪。
+     */
+    private StreamingChatResponseHandler createModelChatResponseHandler(ChatRequest chatRequest,
+                                                                         TraceRunHandle traceRun,
+                                                                         TraceStreamSpan llmSpan) {
+        String sessionId = String.valueOf(chatRequest.getSessionId());
+        return new StreamingChatResponseHandler() {
+
+            private final StringBuilder messageBuffer = new StringBuilder();
+
+            @Override
+            public void onPartialResponse(String partialResponse) {
+                messageBuffer.append(partialResponse);
+                SseMessageUtils.sendContent(sessionId, partialResponse);
+            }
+
+            @Override
+            public void onPartialThinking(PartialThinking partialThinking) {
+                SseMessageUtils.sendReasoning(sessionId, partialThinking.text());
+            }
+
+            @Override
+            public void onCompleteResponse(ChatResponse completeResponse) {
+                try {
+                    String fullMessage = messageBuffer.toString();
+                    if (StringUtils.isNotBlank(fullMessage)) {
+                        chatMessageService.saveChatMessage(
+                            chatRequest.getUserId(),
+                            chatRequest.getSessionId(),
+                            fullMessage,
+                            RoleType.ASSISTANT.getName(),
+                            chatRequest.getModel()
+                        );
+                    } else {
+                        log.warn("普通对话返回空消息,会话:{}", chatRequest.getSessionId());
+                    }
+                    if (llmSpan != null) {
+                        llmSpan.finishSuccess(RagTracePayloadBuilder.streamOutputSummary(fullMessage.length()));
+                    }
+                    finishTraceRun(traceRun, TraceConstants.STATUS_SUCCESS, null);
+                    SseMessageUtils.sendDone(sessionId);
+                } catch (Exception e) {
+                    if (llmSpan != null) {
+                        llmSpan.finishError(e);
+                    }
+                    finishTraceRun(traceRun, TraceConstants.STATUS_ERROR, e);
+                    SseMessageUtils.sendError(sessionId, e.getMessage());
+                    log.error("普通对话完成处理失败", e);
+                } finally {
+                    if (llmSpan != null) {
+                        llmSpan.detach();
+                    }
+                    SseMessageUtils.completeConnection(sessionId);
+                }
+            }
+
+            @Override
+            public void onError(Throwable error) {
+                if (llmSpan != null) {
+                    llmSpan.finishError(error);
+                    llmSpan.detach();
+                }
+                finishTraceRun(traceRun, TraceConstants.STATUS_ERROR, error);
+                SseMessageUtils.sendError(sessionId, error.getMessage());
+                SseMessageUtils.completeConnection(sessionId);
+                log.error("普通对话流式响应失败", error);
+            }
+        };
+    }
+
    /**
     * 创建组合响应处理器 - 同时发送到 SSE 和外部 handler
     *
@@ -811,7 +949,10 @@ public class ChatServiceFacade implements IChatService {
                messageBuffer.append(partialResponse);

                // 2. 发送内容事件到 SSE（前端可通过 SSE 监听）
-                SseMessageUtils.sendContent(sessionId, partialResponse);
+                // 工作流调用时连接归工作流引擎所有, token 由引擎以 [NODE_CHUNK_] 事件推送, 不走聊天协议
+                if (externalHandler == null) {
+                    SseMessageUtils.sendContent(sessionId, partialResponse);
+                }

                // 3. 转发给外部 handler（Workflow 等模块可处理）
                if (externalHandler != null) {
@@ -821,8 +962,10 @@ public class ChatServiceFacade implements IChatService {

            @Override
            public void onPartialThinking(PartialThinking partialThinking) {
-                // 发送推理内容到 SSE（前端通过 reasoning 事件监听）
-                SseMessageUtils.sendReasoning(sessionId, partialThinking.text());
+                // 发送推理内容到 SSE（前端通过 reasoning 事件监听）, 工作流调用时不发送
+                if (externalHandler == null) {
+                    SseMessageUtils.sendReasoning(sessionId, partialThinking.text());
+                }

                // 转发给外部 handler
                if (externalHandler != null) {
@@ -833,11 +976,12 @@ public class ChatServiceFacade implements IChatService {
            @Override
            public void onCompleteResponse(ChatResponse completeResponse) {
                try {
-                    // 1. 发送完成事件
-                    SseMessageUtils.sendDone(sessionId);
-
-                    // 2. 关闭 SSE 连接
-                    SseMessageUtils.completeConnection(sessionId);
+                    // 1&2. 发送完成事件并关闭 SSE 连接
+                    // 工作流调用时流程可能还有后续节点, 连接关闭由工作流引擎统一负责, 此处不能关闭
+                    if (externalHandler == null) {
+                        SseMessageUtils.sendDone(sessionId);
+                        SseMessageUtils.completeConnection(sessionId);
+                    }

                    // 3. 转发给外部 handler
                    if (externalHandler != null) {
@@ -850,8 +994,10 @@ public class ChatServiceFacade implements IChatService {

            @Override
            public void onError(Throwable error) {
-                // 发送错误事件
-                SseMessageUtils.sendError(sessionId, error.getMessage());
+                // 发送错误事件（工作流调用时由工作流引擎统一上报）
+                if (externalHandler == null) {
+                    SseMessageUtils.sendError(sessionId, error.getMessage());
+                }
                log.error("流式响应错误: {}", error.getMessage(), error);

                // 转发给外部 handler
Author	SHA1	Message	Date
ageerle	3c175ba28d	chore: sync pending project changes	2026-08-04 15:23:19 +08:00
ageerle	e34592c3d4	release: prepare v3.1.0	2026-08-04 15:18:57 +08:00
ageerle	fd044a4752	docs: align README badges in one row	2026-07-30 00:22:22 +08:00
ageerle	6e264ad500	feat: streamline workflow orchestration and chat routing Add Zhipu web search integration, remove obsolete workflow nodes and resume handling, and separate model, agent, and workflow chat execution.	2026-07-30 00:15:44 +08:00