"en":"A chat assistant template that integrates information extracted from a knowledge base and web searches to respond to queries. Let's start by setting up your knowledge base in 'Retrieval'!",
"de":"Eine Chat-Assistenten-Vorlage, die Informationen aus einer Wissensdatenbank und Websuchen integriert, um auf Anfragen zu antworten. Beginnen wir mit der Einrichtung Ihrer Wissensdatenbank unter 'Retrieval'!",
"sys_prompt":"Role: You are an Answer Organizer.\nTask: Generate the answer based on the provided content from: User's query, Refined question, Web search result, Retrieval result.\n\nRequirements:\n - Answer should be in markdown format.\n - Answer should include all \n - Do not make thing up when there's no relevant information to user's question. \n",
"temperature":0.1,
"temperatureEnabled":true,
"tools":[],
"topPEnabled":false,
"top_p":0.3,
"user_prompt":"",
"visual_files_var":""
}
},
"upstream":[
"Agent:WildGoatsRule",
"Retrieval:WarmTimesRun"
]
},
"Agent:ThreePathsDecide":{
"downstream":[
"Agent:WildGoatsRule",
"Retrieval:WarmTimesRun"
],
"obj":{
"component_name":"Agent",
"params":{
"delay_after_error":1,
"description":"",
"exception_comment":"",
"exception_default_value":"",
"exception_goto":[],
"exception_method":null,
"frequencyPenaltyEnabled":false,
"frequency_penalty":0.7,
"llm_id":"deepseek-chat@DeepSeek",
"maxTokensEnabled":false,
"max_retries":3,
"max_rounds":1,
"max_tokens":256,
"mcp":[],
"message_history_window_size":12,
"outputs":{
"content":{
"type":"string",
"value":""
}
},
"presencePenaltyEnabled":false,
"presence_penalty":0.4,
"prompts":[
{
"content":"{sys.query}",
"role":"user"
}
],
"sys_prompt":"Role: You are a Question Refinement Agent. Rewrite ambiguous or incomplete user questions to align with knowledge base terminology using conversation history.\n\nExample：\n\nUser: What's RAGFlow?\nAssistant: RAGFlow is xxx.\n\nUser: How to deloy it?\nRefine it: How to deploy RAGFlow?",
"sys_prompt":"Role: You are a Search-Driven Information Agent that answers questions using web search results.\n\nWorkflow:\nKeyword Extraction:\nExtract exactly 3 keywords from the user's question.\n\nKeywords must be:\n✅ Most specific nouns/proper nouns (e.g., \"iPhone 15 Pro\" not \"phone\")\n✅ Core concepts (e.g., \"quantum entanglement\" not \"science thing\")\n✅ Unbiased (no added opinions)\nNever output keywords to users\n\nSearch & Answer:\nUse search tools (TavilySearch, TavilyExtract, Google, Bing, DuckDuckGo, Wikipedia) with the 3 keywords to retrieve results.\nAnswer solely based on search findings, citing sources.\nIf results conflict, prioritize recent (.gov/.edu > forums)\n\nOutput Rules:\n✖️ Never show keywords in final answers\n✖️ Never guess if search yields no results\n✅ Always cite sources using [Source #] notation",
"temperature":0.1,
"temperatureEnabled":true,
"tools":[
{
"component_name":"TavilySearch",
"name":"TavilySearch",
"params":{
"api_key":"",
"days":7,
"exclude_domains":[],
"include_answer":false,
"include_domains":[],
"include_image_descriptions":false,
"include_images":false,
"include_raw_content":true,
"max_results":5,
"outputs":{
"formalized_content":{
"type":"string",
"value":""
},
"json":{
"type":"Array<Object>",
"value":[]
}
},
"query":"sys.query",
"search_depth":"basic",
"topic":"general"
}
},
{
"component_name":"TavilyExtract",
"name":"TavilyExtract",
"params":{
"api_key":""
}
},
{
"component_name":"Google",
"name":"Google",
"params":{
"api_key":"",
"country":"us",
"language":"en"
}
},
{
"component_name":"Bing",
"name":"Bing",
"params":{
"api_key":"YOUR_API_KEY (obtained from https://www.microsoft.com/en-us/bing/apis/bing-web-search-api)",
"channel":"Webpages",
"country":"CH",
"language":"en",
"top_n":10
}
},
{
"component_name":"DuckDuckGo",
"name":"DuckDuckGo",
"params":{
"channel":"text",
"top_n":10
}
},
{
"component_name":"Wikipedia",
"name":"Wikipedia",
"params":{
"language":"en",
"top_n":10
}
}
],
"topPEnabled":false,
"top_p":0.3,
"user_prompt":"",
"visual_files_var":""
}
},
"upstream":[
"Agent:ThreePathsDecide"
]
},
"Message:ShaggyRingsCrash":{
"downstream":[],
"obj":{
"component_name":"Message",
"params":{
"content":[
"{Agent:SmartSchoolsCross@content}"
]
}
},
"upstream":[
"Agent:SmartSchoolsCross"
]
},
"Retrieval:WarmTimesRun":{
"downstream":[
"Agent:SmartSchoolsCross"
],
"obj":{
"component_name":"Retrieval",
"params":{
"cross_languages":[],
"empty_response":"",
"kb_ids":[],
"keywords_similarity_weight":0.7,
"outputs":{
"formalized_content":{
"type":"string",
"value":""
}
},
"query":"Agent:ThreePathsDecide@content",
"rerank_id":"",
"similarity_threshold":0.2,
"top_k":1024,
"top_n":8,
"use_kg":false
}
},
"upstream":[
"Agent:ThreePathsDecide"
]
},
"begin":{
"downstream":[
"Agent:ThreePathsDecide"
],
"obj":{
"component_name":"Begin",
"params":{
"enablePrologue":true,
"inputs":{},
"mode":"conversational",
"prologue":"Hi! I'm your web search assistant. What do you want to search today?"
"prologue":"Hi! I'm your web search assistant. What do you want to search today?"
},
"label":"Begin",
"name":"begin"
},
"dragging":false,
"id":"begin",
"measured":{
"height":48,
"width":200
},
"position":{
"x":32.79251060693639,
"y":209.67921278359827
},
"selected":false,
"sourcePosition":"left",
"targetPosition":"right",
"type":"beginNode"
},
{
"data":{
"form":{
"delay_after_error":1,
"description":"",
"exception_comment":"",
"exception_default_value":"",
"exception_goto":[],
"exception_method":null,
"frequencyPenaltyEnabled":false,
"frequency_penalty":0.7,
"llm_id":"deepseek-chat@DeepSeek",
"maxTokensEnabled":false,
"max_retries":3,
"max_rounds":1,
"max_tokens":256,
"mcp":[],
"message_history_window_size":12,
"outputs":{
"content":{
"type":"string",
"value":""
}
},
"presencePenaltyEnabled":false,
"presence_penalty":0.4,
"prompts":[
{
"content":"{sys.query}",
"role":"user"
}
],
"sys_prompt":"Role: You are a Question Refinement Agent. Rewrite ambiguous or incomplete user questions to align with knowledge base terminology using conversation history.\n\nExample：\n\nUser: What's RAGFlow?\nAssistant: RAGFlow is xxx.\n\nUser: How to deloy it?\nRefine it: How to deploy RAGFlow?",
"sys_prompt":"Role: You are a Search-Driven Information Agent that answers questions using web search results.\n\nWorkflow:\nKeyword Extraction:\nExtract exactly 3 keywords from the user's question.\n\nKeywords must be:\n✅ Most specific nouns/proper nouns (e.g., \"iPhone 15 Pro\" not \"phone\")\n✅ Core concepts (e.g., \"quantum entanglement\" not \"science thing\")\n✅ Unbiased (no added opinions)\nNever output keywords to users\n\nSearch & Answer:\nUse search tools (TavilySearch, TavilyExtract, Google, Bing, DuckDuckGo, Wikipedia) with the 3 keywords to retrieve results.\nAnswer solely based on search findings, citing sources.\nIf results conflict, prioritize recent (.gov/.edu > forums)\n\nOutput Rules:\n✖️ Never show keywords in final answers\n✖️ Never guess if search yields no results\n✅ Always cite sources using [Source #] notation",
"temperature":0.1,
"temperatureEnabled":true,
"tools":[
{
"component_name":"TavilySearch",
"name":"TavilySearch",
"params":{
"api_key":"",
"days":7,
"exclude_domains":[],
"include_answer":false,
"include_domains":[],
"include_image_descriptions":false,
"include_images":false,
"include_raw_content":true,
"max_results":5,
"outputs":{
"formalized_content":{
"type":"string",
"value":""
},
"json":{
"type":"Array<Object>",
"value":[]
}
},
"query":"sys.query",
"search_depth":"basic",
"topic":"general"
}
},
{
"component_name":"TavilyExtract",
"name":"TavilyExtract",
"params":{
"api_key":""
}
},
{
"component_name":"Google",
"name":"Google",
"params":{
"api_key":"",
"country":"us",
"language":"en"
}
},
{
"component_name":"Bing",
"name":"Bing",
"params":{
"api_key":"YOUR_API_KEY (obtained from https://www.microsoft.com/en-us/bing/apis/bing-web-search-api)",
"sys_prompt":"Role: You are an Answer Organizer.\nTask: Generate the answer based on the provided content from: User's query, Refined question, Web search result, Retrieval result.\n\nRequirements:\n - Answer should be in markdown format.\n - Answer should include all \n - Do not make thing up when there's no relevant information to user's question. \n",
"temperature":0.1,
"temperatureEnabled":true,
"tools":[],
"topPEnabled":false,
"top_p":0.3,
"user_prompt":"",
"visual_files_var":""
},
"label":"Agent",
"name":"Answer Organizer"
},
"dragging":false,
"id":"Agent:SmartSchoolsCross",
"measured":{
"height":84,
"width":200
},
"position":{
"x":1134.5321493898284,
"y":221.46972754101765
},
"selected":false,
"sourcePosition":"right",
"targetPosition":"left",
"type":"agentNode"
},
{
"data":{
"form":{
"content":[
"{Agent:SmartSchoolsCross@content}"
]
},
"label":"Message",
"name":"Answer"
},
"dragging":false,
"id":"Message:ShaggyRingsCrash",
"measured":{
"height":56,
"width":200
},
"position":{
"x":1437.758553651028,
"y":235.45081267288185
},
"selected":false,
"sourcePosition":"right",
"targetPosition":"left",
"type":"messageNode"
},
{
"data":{
"form":{
"text":"This Agent rewrites your question for better search & retrieval results."
},
"label":"Note",
"name":"Note: Refine Question"
},
"dragHandle":".note-drag-handle",
"id":"Note:BetterCupsBow",
"measured":{
"height":136,
"width":244
},
"position":{
"x":270,
"y":390
},
"selected":false,
"sourcePosition":"right",
"targetPosition":"left",
"type":"noteNode"
},
{
"data":{
"form":{
"text":"This Agent answers questions using web search results."
},
"label":"Note",
"name":"Note: Search Agent"
},
"dragHandle":".note-drag-handle",
"dragging":false,
"id":"Note:OddGoatsBeg",
"measured":{
"height":136,
"width":244
},
"position":{
"x":689.3401860180043,
"y":-204.46057070562227
},
"selected":false,
"sourcePosition":"right",
"targetPosition":"left",
"type":"noteNode"
},
{
"data":{
"form":{
"text":"This Agents generates the answer based on the provided content from: User's query, Refined question, Web search result, Retrieval result."
},
"label":"Note",
"name":"Note: Answer Organizer"
},
"dragHandle":".note-drag-handle",
"dragging":false,
"height":188,
"id":"Note:SlowBottlesHope",
"measured":{
"height":188,
"width":251
},
"position":{
"x":1152.1929528629184,
"y":375.08305219772546
},
"resizing":false,
"selected":false,
"sourcePosition":"right",
"targetPosition":"left",
"type":"noteNode",
"width":251
},
{
"data":{
"form":{
"description":"This is an agent for a specific task.",
"user_prompt":"This is the order you need to send to the agent."
},
"label":"Tool",
"name":"flow.tool_0"
},
"dragging":false,
"id":"Tool:TrueCrewsTake",
"measured":{
"height":228,
"width":200
},
"position":{
"x":642.9703031510875,
"y":144.80253344921545
},
"selected":false,
"sourcePosition":"right",
"targetPosition":"left",
"type":"toolNode"
},
{
"data":{
"form":{
"text":"This is a chat assistant template that integrates information extracted from a knowledge base and web searches to respond to queries. Let's start by setting up your knowledge base in 'Retrieval'!"
