通过全局变量和变量更新，将提问的方式收集起来

[anomone]

例行检查

• 我已确认目前没有类似 features
• 我已确认我已升级到最新版本
• 我已完整查看过项目 README，已确定现有版本无法满足需求
• 我理解并愿意跟进此 features，协助测试和提供反馈
• 我理解并认可上述内容，并理解项目维护者精力有限，不遵循规则的 features 可能会被无视或直接关闭

功能描述
能否让全局变量以提问的方式跟用户互动，最终收集数据组装成提问 再发送给ai回答
应用场景
客服
相关示例
以旅游智能客服举例
智能客服会收集三个信息，“请问您的旅游天数？”，“请问您的旅游人数？”，“请问您的旅游目的地是哪里？”
如果用全局变量去收集太生硬了。

效果用百度智能客服举例，感觉稍微改改就能实现智能客服这个方向了（或者有办法但是我能力有限没想到）

[lijiajun1997]

从客户体验来说还不如用全局变量收集或者一次性问完呢，其实直接放在prompt提问里也可以。
按你的要求也不是不行，就是比较复杂而且浪费时间：

[lijiajun1997]

{
"nodes": [
{
"nodeId": "userGuide",
"name": "系统配置",
"intro": "可以配置应用的系统参数",
"avatar": "core/workflow/template/systemConfig",
"flowNodeType": "userGuide",
"position": {
"x": 531.2422736065552,
"y": -486.7611729549753
},
"version": "481",
"inputs": [
{
"key": "welcomeText",
"renderTypeList": [
"hidden"
],
"valueType": "string",
"label": "core.app.Welcome Text",
"value": ""
},
{
"key": "variables",
"renderTypeList": [
"hidden"
],
"valueType": "any",
"label": "core.app.Chat Variable",
"value": []
},
{
"key": "questionGuide",
"valueType": "boolean",
"renderTypeList": [
"hidden"
],
"label": "core.app.Question Guide",
"value": false
},
{
"key": "tts",
"renderTypeList": [
"hidden"
],
"valueType": "any",
"label": "",
"value": {
"type": "web"
}
},
{
"key": "whisper",
"renderTypeList": [
"hidden"
],
"valueType": "any",
"label": "",
"value": {
"open": false,
"autoSend": false,
"autoTTSResponse": false
}
},
{
"key": "scheduleTrigger",
"renderTypeList": [
"hidden"
],
"valueType": "any",
"label": "",
"value": null
}
],
"outputs": []
},
{
"nodeId": "448745",
"name": "流程开始",
"intro": "",
"avatar": "core/workflow/template/workflowStart",
"flowNodeType": "workflowStart",
"position": {
"x": 683.6217608535718,
"y": 302.3230081757545
},
"version": "481",
"inputs": [
{
"key": "userChatInput",
"renderTypeList": [
"reference",
"textarea"
],
"valueType": "string",
"label": "用户问题",
"required": true,
"toolDescription": "用户问题"
}
],
"outputs": [
{
"id": "userChatInput",
"key": "userChatInput",
"label": "common:core.module.input.label.user question",
"type": "static",
"valueType": "string"
}
]
},
{
"nodeId": "y3B4gjlLbTz7",
"name": "变量更新",
"intro": "可以更新指定节点的输出值或更新全局变量",
"avatar": "core/workflow/template/variableUpdate",
"flowNodeType": "variableUpdate",
"showStatus": false,
"position": {
"x": 2681.5762224917707,
"y": -302.4419527946851
},
"version": "481",
"inputs": [
{
"key": "updateList",
"valueType": "any",
"label": "",
"renderTypeList": [
"hidden"
],
"value": [
{
"variable": [
"VARIABLE_NODE_ID",
"target"
],
"value": [
"owwL79CHWdDD",
"gi9BFSpLY1nj"
],
"valueType": "string",
"renderType": "reference"
}
]
}
],
"outputs": []
},
{
"nodeId": "sFeHXAx2b2Vs",
"name": "AI 对话",
"intro": "AI 大模型对话",
"avatar": "core/workflow/template/aiChat",
"flowNodeType": "chatNode",
"showStatus": true,
"position": {
"x": 3038.5121928842154,
"y": 59.09803763616847
},
"version": "481",
"inputs": [
{
"key": "model",
"renderTypeList": [
"settingLLMModel",
"reference"
],
"label": "core.module.input.label.aiModel",
"valueType": "string",
"value": "gpt-4o-mini"
},
{
"key": "temperature",
"renderTypeList": [
"hidden"
],
"label": "",
"value": 0,
"valueType": "number"
},
{
"key": "maxToken",
"renderTypeList": [
"hidden"
],
"label": "",
"value": 2000,
"valueType": "number"
},
{
"key": "isResponseAnswerText",
"renderTypeList": [
"hidden"
],
"label": "",
"value": true,
"valueType": "boolean"
},
{
"key": "quoteTemplate",
"renderTypeList": [
"hidden"
],
"label": "",
"valueType": "string"
},
{
"key": "quotePrompt",
"renderTypeList": [
"hidden"
],
"label": "",
"valueType": "string"
},
{
"key": "aiChatVision",
"renderTypeList": [
"hidden"
],
"label": "",
"valueType": "boolean",
"value": true
},
{
"key": "systemPrompt",
"renderTypeList": [
"textarea",
"reference"
],
"max": 3000,
"valueType": "string",
"label": "core.ai.Prompt",
"description": "core.app.tip.chatNodeSystemPromptTip",
"placeholder": "core.app.tip.chatNodeSystemPromptTip",
"value": "用户目的地是：{{$VARIABLE_NODE_ID.target$}}\n用户的旅行天数是{{$VARIABLE_NODE_ID.days$}}\n如果上述信息为空，你需要通过提问获取信息\n"
},
{
"key": "history",
"renderTypeList": [
"numberInput",
"reference"
],
"valueType": "chatHistory",
"label": "core.module.input.label.chat history",
"description": "最多携带多少轮对话记录",
"required": true,
"min": 0,
"max": 50,
"value": 0
},
{
"key": "quoteQA",
"renderTypeList": [
"settingDatasetQuotePrompt"
],
"label": "",
"debugLabel": "知识库引用",
"description": "",
"valueType": "datasetQuote"
},
{
"key": "stringQuoteText",
"renderTypeList": [
"reference",
"textarea"
],
"label": "app:document_quote",
"debugLabel": "app:document_quote",
"description": "app:document_quote_tip",
"valueType": "string"
},
{
"key": "userChatInput",
"renderTypeList": [
"reference",
"textarea"
],
"valueType": "string",
"label": "用户问题",
"required": true,
"toolDescription": "用户问题",
"value": [
"448745",
"userChatInput"
]
}
],
"outputs": [
{
"id": "history",
"key": "history",
"required": true,
"label": "core.module.output.label.New context",
"description": "core.module.output.description.New context",
"valueType": "chatHistory",
"type": "static"
},
{
"id": "answerText",
"key": "answerText",
"required": true,
"label": "core.module.output.label.Ai response content",
"description": "core.module.output.description.Ai response content",
"valueType": "string",
"type": "static"
}
]
},
{
"nodeId": "owwL79CHWdDD",
"name": "文本内容提取",
"intro": "可从文本中提取指定的数据，例如：sql语句、搜索关键词、代码等",
"avatar": "core/workflow/template/extractJson",
"flowNodeType": "contentExtract",
"showStatus": true,
"position": {
"x": 987.0370712605812,
"y": 90.71983265598982
},
"version": "481",
"inputs": [
{
"key": "model",
"renderTypeList": [
"selectLLMModel",
"reference"
],
"label": "core.module.input.label.aiModel",
"required": true,
"valueType": "string",
"llmModelType": "extractFields",
"value": "gpt-4o-mini"
},
{
"key": "description",
"renderTypeList": [
"textarea",
"reference"
],
"valueType": "string",
"label": "提取要求描述",
"description": "给AI一些对应的背景知识或要求描述，引导AI更好的完成任务。\n该输入框可使用全局变量。",
"placeholder": "例如: \n1. 当前时间为: {{cTime}}。你是一个实验室预约助手，你的任务是帮助用户预约实验室，从文本中获取对应的预约信息。\n2. 你是谷歌搜索助手，需要从文本中提取出合适的搜索词。",
"value": "用户目的地是：{{$VARIABLE_NODE_ID.target$}}\n用户的旅行天数是{{$VARIABLE_NODE_ID.days$}}\n如果需要从输入的文本中提取信息替代上面的信息"
},
{
"key": "history",
"renderTypeList": [
"numberInput",
"reference"
],
"valueType": "chatHistory",
"label": "core.module.input.label.chat history",
"description": "最多携带多少轮对话记录",
"required": true,
"min": 0,
"max": 50,
"value": 6
},
{
"key": "content",
"renderTypeList": [
"reference",
"textarea"
],
"label": "需要提取的文本",
"required": true,
"valueType": "string",
"toolDescription": "需要检索的内容",
"value": [
"448745",
"userChatInput"
]
},
{
"key": "extractKeys",
"renderTypeList": [
"custom"
],
"label": "",
"valueType": "any",
"description": "由 '描述' 和 'key' 组成一个目标字段，可提取多个目标字段",
"value": [
{
"valueType": "string",
"required": false,
"defaultValue": "",
"desc": "用户旅游的目的地",
"key": "target",
"enum": ""
},
{
"valueType": "string",
"required": false,
"defaultValue": "",
"desc": "旅游的天数",
"key": "days",
"enum": ""
}
]
}
],
"outputs": [
{
"id": "success",
"key": "success",
"label": "字段完全提取",
"required": true,
"description": "提取字段全部填充时返回 true （模型提取或使用默认值均属于成功）",
"valueType": "boolean",
"type": "static"
},
{
"id": "fields",
"key": "fields",
"label": "完整提取结果",
"required": true,
"description": "一个 JSON 字符串，例如：{"name:":"YY","Time":"2023/7/2 18:00"}",
"valueType": "string",
"type": "static"
},
{
"id": "gi9BFSpLY1nj",
"key": "target",
"label": "提取结果-用户旅游的目的地",
"valueType": "string",
"type": "static"
},
{
"id": "uLiCRPjyItaX",
"key": "days",
"label": "提取结果-旅游的天数",
"valueType": "string",
"type": "static"
}
]
},
{
"nodeId": "mi9g6NPe5Qri",
"name": "判断器",
"intro": "根据一定的条件，执行不同的分支。",
"avatar": "core/workflow/template/ifelse",
"flowNodeType": "ifElseNode",
"showStatus": true,
"position": {
"x": 1631.3534237880635,
"y": -425.87340444891225
},
"version": "481",
"inputs": [
{
"key": "ifElseList",
"renderTypeList": [
"hidden"
],
"valueType": "any",
"label": "",
"value": [
{
"condition": "AND",
"list": [
{
"variable": [
"owwL79CHWdDD",
"gi9BFSpLY1nj"
],
"condition": "isNotEmpty"
}
]
}
]
}
],
"outputs": [
{
"id": "ifElseResult",
"key": "ifElseResult",
"label": "判断结果",
"valueType": "string",
"type": "static"
}
]
},
{
"nodeId": "h1TxF3QNcush",
"name": "判断器#2",
"intro": "根据一定的条件，执行不同的分支。",
"avatar": "core/workflow/template/ifelse",
"flowNodeType": "ifElseNode",
"showStatus": true,
"position": {
"x": 1585.912398395383,
"y": 161.88378154359145
},
"version": "481",
"inputs": [
{
"key": "ifElseList",
"renderTypeList": [
"hidden"
],
"valueType": "any",
"label": "",
"value": [
{
"condition": "AND",
"list": [
{
"variable": [
"owwL79CHWdDD",
"uLiCRPjyItaX"
],
"condition": "isNotEmpty"
}
]
}
]
}
],
"outputs": [
{
"id": "ifElseResult",
"key": "ifElseResult",
"label": "判断结果",
"valueType": "string",
"type": "static"
}
]
},
{
"nodeId": "xaX3RUAk6tud",
"name": "变量更新#2",
"intro": "可以更新指定节点的输出值或更新全局变量",
"avatar": "core/workflow/template/variableUpdate",
"flowNodeType": "variableUpdate",
"showStatus": false,
"position": {
"x": 2461.01421150823,
"y": 88.31608974152292
},
"version": "481",
"inputs": [
{
"key": "updateList",
"valueType": "any",
"label": "",
"renderTypeList": [
"hidden"
],
"value": [
{
"variable": [
"VARIABLE_NODE_ID",
"days"
],
"value": [
"owwL79CHWdDD",
"uLiCRPjyItaX"
],
"valueType": "string",
"renderType": "reference"
}
]
}
],
"outputs": []
}
],
"edges": [
{
"source": "448745",
"target": "owwL79CHWdDD",
"sourceHandle": "448745-source-right",
"targetHandle": "owwL79CHWdDD-target-left"
},
{
"source": "owwL79CHWdDD",
"target": "mi9g6NPe5Qri",
"sourceHandle": "owwL79CHWdDD-source-right",
"targetHandle": "mi9g6NPe5Qri-target-left"
},
{
"source": "mi9g6NPe5Qri",
"target": "y3B4gjlLbTz7",
"sourceHandle": "mi9g6NPe5Qri-source-IF",
"targetHandle": "y3B4gjlLbTz7-target-left"
},
{
"source": "owwL79CHWdDD",
"target": "h1TxF3QNcush",
"sourceHandle": "owwL79CHWdDD-source-right",
"targetHandle": "h1TxF3QNcush-target-left"
},
{
"source": "h1TxF3QNcush",
"target": "xaX3RUAk6tud",
"sourceHandle": "h1TxF3QNcush-source-IF",
"targetHandle": "xaX3RUAk6tud-target-left"
},
{
"source": "xaX3RUAk6tud",
"target": "sFeHXAx2b2Vs",
"sourceHandle": "xaX3RUAk6tud-source-right",
"targetHandle": "sFeHXAx2b2Vs-target-left"
},
{
"source": "h1TxF3QNcush",
"target": "sFeHXAx2b2Vs",
"sourceHandle": "h1TxF3QNcush-source-ELSE",
"targetHandle": "sFeHXAx2b2Vs-target-left"
}
],
"chatConfig": {
"scheduledTriggerConfig": {
"cronString": "",
"timezone": "Asia/Shanghai",
"defaultPrompt": ""
},
"variables": [
{
"id": "7gm1hc",
"key": "target",
"label": "target",
"type": "custom",
"required": false,
"maxLen": 50,
"enums": [
{
"value": ""
}
],
"icon": "core/app/variable/input"
},
{
"id": "u58hs7",
"key": "days",
"label": "days",
"type": "custom",
"required": false,
"maxLen": 50,
"enums": [
{
"value": ""
}
],
"icon": "core/app/variable/input"
}
]
}
}

[anomone]

从客户体验来说还不如用全局变量收集或者一次性问完呢，其实直接放在prompt提问里也可以。 按你的要求也不是不行，就是比较复杂而且浪费时间：

受您的启发，通过判断器、变量更新 和 代码执行已经实现了，非常感谢。
就是对话开场白不能分成两段，流程发起总是用户，不能是ai方，开场白效果有点勉强

加了这一段

效果相当Nice,fastgpt牛逼

[c121914yu]

转成讨论了哈～ 这个需求挺经典的