Custom1768595176788.yml

flowMeta:
  name: Custom1768595176788
  description: ''
  avatarIcon: https://oss-beijing-m8.openstorage.cn/SparkBotProd/icon/common/emojiitem_00_10@2x.png
  avatarColor: '#FFEAD5'
  advancedConfig: '{"prologue":{"enabled":true,"inputExample":["","",""]},"textToSpeech":{"enabled":true,"vcn_cn":"x5_lingxiaotang_flow","vcn_en":""}}'
  dslVersion: v1
flowData:
  nodes:
  - id: node-start::d61b0f71-87ee-475e-93ba-f1607f0ce783
    dragging: false
    selected: false
    width: 362
    height: 82
    position:
      x: -25.109019607843152
      y: 521.7086666666667
    positionAbsolute:
      x: -25.109019607843152
      y: 521.7086666666667
    type: custom
    data:
      allowInputReference: false
      allowOutputReference: true
      label: 开始
      status: ''
      nodeMeta:
        aliasName: 开始节点
        nodeType: 基础节点
      inputs: [
        ]
      outputs:
      - id: 0918514b-72a8-4646-8dd9-ff4a8fc26d44
        name: AGENT_USER_INPUT
        nameErrMsg: ''
        schema:
          type: string
          default: 用户本轮对话输入内容
        required: true
        deleteDisabled: true
      nodeParam: {
        }
      icon: https://oss-beijing-m8.openstorage.cn/pro-bucket/sparkBot/common/workflow/icon/start-node-icon.png
      description: 工作流的开启节点，用于定义流程调用所需的业务变量信息。
      updatable: false
  - id: node-end::cda617af-551e-462e-b3b8-3bb9a041bf88
    dragging: false
    selected: true
    width: 362
    height: 82
    position:
      x: 886.8833333333332
      y: 343.91588235294114
    positionAbsolute:
      x: 886.8833333333332
      y: 343.91588235294114
    type: custom
    data:
      allowInputReference: true
      allowOutputReference: false
      label: 结束
      references: [
        ]
      status: ''
      nodeMeta:
        aliasName: 结束节点
        nodeType: 基础节点
      inputs:
      - id: 82de2b42-a059-4c98-bffb-b6b4800fcac9
        name: output
        nameErrMsg: ''
        schema:
          type: string
          value:
            type: ref
            content:
              id: 81b74b7c-7b33-4b2c-bbf2-4eaa9cc5e535
              nodeId: spark-llm::05792386-f413-4c13-b87a-e2fc96be8432
              name: output
            contentErrMsg: ''
        fileType: ''
      outputs: [
        ]
      nodeParam:
        template: '{{output}}'
        streamOutput: true
        outputMode: 1
        templateErrMsg: ''
      icon: https://oss-beijing-m8.openstorage.cn/pro-bucket/sparkBot/common/workflow/icon/end-node-icon.png
      description: 工作流的结束节点，用于输出工作流运行后的最终结果。
      updatable: false
  - id: spark-llm::05792386-f413-4c13-b87a-e2fc96be8432
    selected: false
    width: 362
    height: 132
    position:
      x: 577.2387173354679
      y: 569.6257981625986
    type: custom
    data:
      allowInputReference: true
      allowOutputReference: true
      label: Large Model_1
      labelEdit: false
      references: [
        ]
      status: ''
      nodeMeta:
        nodeType: Basic Node
        aliasName: 大模型
      inputs:
      - id: e9a63719-c01d-4d8d-98f9-2c3a649dbb91
        name: input
        nameErrMsg: ''
        schema:
          type: string
          value:
            type: ref
            content:
              id: 0918514b-72a8-4646-8dd9-ff4a8fc26d44
              nodeId: node-start::d61b0f71-87ee-475e-93ba-f1607f0ce783
              name: AGENT_USER_INPUT
            contentErrMsg: ''
        fileType: ''
      outputs:
      - id: 81b74b7c-7b33-4b2c-bbf2-4eaa9cc5e535
        name: output
        nameErrMsg: ''
        schema:
          type: string
          default: ''
      nodeParam:
        maxTokens: 2048
        topK: 4
        auditing: default
        domain: deepseek-chat
        template: ~{{input}}
        model: spark
        serviceId: deepseek-chat
        respFormat: 0
        llmId: 593066024
        url: https://api.deepseek.com/v1/chat/completions
        appId: 680ab54f
        uid: af4d88d6-7c28-4834-bf4e-cce81afa6231
        enableChatHistoryV2:
          isEnabled: false
          rounds: 1
        templateErrMsg: ''
        modelId: 1
        isThink: false
        multiMode: false
        modelName: demo
        modelEnabled: true
        llmIdErrMsg: ''
        source: openai
        extraParams:
          temperature: 1
        systemTemplate: |+
          ~You are an AI assistant that helps convert technical documentation into structured courses for junior developers. Your task is to take raw documentation and deconstruct it into the following elements, then output the results in JSON format:










          1. **Course Outline**: Identify the main topics or modules in the documentation. Each module should cover a distinct concept or section.




          2. **Lessons**: Break down each module into individual lessons that are small, manageable, and focused on specific concepts or steps. Include explanations and relevant code examples or snippets for each lesson.




          3. **Key Concepts**: Identify key concepts, terminology, or functions that need to be defined for a junior developer to understand them better.










          The JSON format should look like this:










          {




            "course_title": "[Course Title]",




            "course_description": "[Brief Description of the Course]",




            "modules": [




              {




                "module_title": "[Module Title]",




                "module_description": "[Module Description]",




                "lessons": [




                  {




                    "lesson_title": "[Lesson Title]",




                    "lesson_content": "[Detailed Lesson Explanation]",




                    "code_example": "[Code Example]",




                    "key_concepts": ["[Concept 1]", "[Concept 2]", ...]




                  },




                  ...




                ]




              },




              ...




            ]




          }










          The generated course should help junior developers easily follow along, grasping each concept before moving on to the next. Keep the language simple and avoid overwhelming technical jargon.


          output strictly only the json format, and no additional text whatsoever



      icon: https://oss-beijing-m8.openstorage.cn/pro-bucket/sparkBot/common/workflow/icon/largeModelIcon.png
      description: Based on the input prompt, the selected large language model will
        be invoked to respond accordingly.
      updatable: false
  edges:
  - source: node-start::d61b0f71-87ee-475e-93ba-f1607f0ce783
    target: spark-llm::05792386-f413-4c13-b87a-e2fc96be8432
    type: customEdge
    id: reactflow__edge-node-start::d61b0f71-87ee-475e-93ba-f1607f0ce783-spark-llm::05792386-f413-4c13-b87a-e2fc96be8432
    markerEnd:
      type: arrow
      color: '#6356EA'
    data:
      edgeType: curve
  - source: spark-llm::05792386-f413-4c13-b87a-e2fc96be8432
    target: node-end::cda617af-551e-462e-b3b8-3bb9a041bf88
    type: customEdge
    id: reactflow__edge-spark-llm::05792386-f413-4c13-b87a-e2fc96be8432-node-end::cda617af-551e-462e-b3b8-3bb9a041bf88
    markerEnd:
      type: arrow
      color: '#6356EA'
    data:
      edgeType: curve