This workflow follows the Agent → Chainllm recipe pattern — see all workflows that pair these two integrations.
The workflow JSON
Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →
{
"name": "crawl4 ai",
"nodes": [
{
"parameters": {
"options": {}
},
"id": "2e66873a-c0a0-44d3-83e9-6b993428aa44",
"name": "Loop Over Items",
"type": "n8n-nodes-base.splitInBatches",
"typeVersion": 3,
"position": [
280,
1180
]
},
{
"parameters": {
"jsonMode": "expressionData",
"jsonData": "={{ $('HTTP Request8').item.json.results[0].markdown }}",
"options": {
"metadata": {
"metadataValues": [
{
"name": "url",
"value": "={{ $('HTTP Request8').item.json.results[0].url }}"
},
{
"name": "query",
"value": "={{ $('extract url and query').item.json.query }}"
},
{}
]
}
}
},
"id": "af4a8f64-e866-4f13-8d95-02ae647bbb38",
"name": "Default Data Loader",
"type": "@n8n/n8n-nodes-langchain.documentDefaultDataLoader",
"typeVersion": 1,
"position": [
1400,
1460
]
},
{
"parameters": {
"chunkSize": 700,
"chunkOverlap": 100
},
"id": "4d7d1fa9-08ed-4a08-b733-97c84899d1f2",
"name": "Character Text Splitter",
"type": "@n8n/n8n-nodes-langchain.textSplitterCharacterTextSplitter",
"typeVersion": 1,
"position": [
1540,
1600
]
},
{
"parameters": {
"mode": "insert",
"tableName": {
"__rl": true,
"value": "documents",
"mode": "list",
"cachedResultName": "documents"
},
"options": {
"queryName": "match_documents"
}
},
"id": "effd12a0-b7fe-442b-a3b3-ccbe931c4106",
"name": "Supabase Vector Store",
"type": "@n8n/n8n-nodes-langchain.vectorStoreSupabase",
"typeVersion": 1,
"position": [
1300,
1200
],
"alwaysOutputData": true,
"retryOnFail": false,
"credentials": {
"supabaseApi": {
"name": "<your credential>"
}
},
"onError": "continueRegularOutput"
},
{
"parameters": {
"content": "## 1.1 Set the search parameters\n\nIn the first step the search parameters can be set.\nThe LLM will automatically generate different search request variations\n\n- use **Set search request** to set the search parameter\n- you can prompt the amount of variations in the **Get different search variations** node change the amount: `1\u20132 Strings` to your liking\n\n",
"height": 234,
"width": 590,
"color": 7
},
"id": "b5a6ca6f-01dd-4f07-a570-4cd08bc6b4e0",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
-1680,
900
]
},
{
"parameters": {
"fieldToSplitOut": "results[0].markdown",
"options": {
"destinationFieldName": "documents"
}
},
"type": "n8n-nodes-base.splitOut",
"typeVersion": 1,
"position": [
740,
1200
],
"id": "686f4f2c-adc4-44b3-b4e1-83afd7752c95",
"name": "Split Out1"
},
{
"parameters": {
"model": "mxbai-embed-large:latest"
},
"type": "@n8n/n8n-nodes-langchain.embeddingsOllama",
"typeVersion": 1,
"position": [
1220,
1460
],
"id": "b4dbcaa5-8677-4a8a-a58b-36d7085dfaca",
"name": "Embeddings Ollama",
"credentials": {
"ollamaApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"schemaType": "manual",
"inputSchema": "{\n \"type\": \"object\",\n \"properties\": {\n \"variations\": {\n \"type\": \"array\",\n \"items\": { \"type\": \"string\" }\n }\n },\n \"required\": [\"variations\"]\n}"
},
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"typeVersion": 1.2,
"position": [
-1120,
1360
],
"id": "51a760e1-82e4-4f1a-86c8-c857ff0b8cce",
"name": "Structured Output Parser"
},
{
"parameters": {
"url": "http://127.0.0.1:8080/search",
"sendQuery": true,
"specifyQuery": "=keypair",
"queryParameters": {
"parameters": [
{
"name": "=q",
"value": "={{ $json.variations }}"
},
{
"name": "format",
"value": "json"
}
]
},
"options": {
"batching": {
"batch": {
"batchSize": 1
}
}
}
},
"type": "n8n-nodes-base.httpRequest",
"typeVersion": 4.2,
"position": [
-580,
1180
],
"id": "88256f43-e936-403a-a92a-32b74aeaa71d",
"name": "searxng"
},
{
"parameters": {
"modelName": "models/gemini-2.0-flash-001",
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
"typeVersion": 1,
"position": [
-1320,
1360
],
"id": "9a82a3c2-e254-456c-8ad8-9ea00902c690",
"name": "Google Gemini Chat Model2",
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"fieldToSplitOut": "output.variations",
"options": {
"destinationFieldName": "variations"
}
},
"type": "n8n-nodes-base.splitOut",
"typeVersion": 1,
"position": [
-980,
1180
],
"id": "a74618c5-1308-4bd2-aaae-02c8e909f8cd",
"name": "Split Out2"
},
{
"parameters": {
"fieldToSplitOut": "results",
"include": "selectedOtherFields",
"fieldsToInclude": "query",
"options": {
"destinationFieldName": "results"
}
},
"type": "n8n-nodes-base.splitOut",
"typeVersion": 1,
"position": [
-400,
1180
],
"id": "d5bd650f-9cd0-4d62-a984-9cecc3d74ebd",
"name": "Split Out3"
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "14facf3f-b95f-43db-8ffa-cf3ded5ba5e0",
"name": "topic",
"value": "e-mobility trends",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
-1500,
1180
],
"id": "0bce2b72-154d-49c9-9af2-f258657b990e",
"name": "Set search request"
},
{
"parameters": {
"promptType": "define",
"text": "=Du bist ein Assistant, der mir hilft, f\u00fcr eine Suchanfrage alle sinnvollen Varianten (Synonyme, englische/ deutsche Schreibweisen, Zusammen\u2011/Getrenntschreibung) auszuspielen. \nEingabe: ein Thema {{ $json.topic }}. \nAusgabe: **nur** ein g\u00fcltiges JSON\u2011Objekt mit genau einem Feld \"variations\", das eine Liste von 1\u20132 Strings enth\u00e4lt.\nBeispiel\u2011Schema:\n{\n \"type\": \"object\",\n \"properties\": {\n \"variations\": {\n \"type\": \"array\",\n \"items\": { \"type\": \"string\" }\n }\n },\n \"required\": [\"variations\"]\n}\n\n**Keine** zus\u00e4tzlichen Erkl\u00e4rungen oder Floskeln, **nur** das JSON.\n",
"hasOutputParser": true
},
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"typeVersion": 1.6,
"position": [
-1320,
1180
],
"id": "f208cc0b-5066-4b62-b9ec-f87d6899b4a8",
"name": "Get different search variations",
"alwaysOutputData": true,
"retryOnFail": true,
"waitBetweenTries": 5000,
"onError": "continueRegularOutput"
},
{
"parameters": {
"content": "## 1.2 Search for URLs with Searxng\n\nIt is necessary to run the searxng docker: https://github.com/searxng/searxng-docker\nto run on windows you also need to run the docker desktop app.\n\n1. edit the settings.yml:\n\n```\nuse_default_settings: true\n\nserver:\n # change this key\n secret_key: \"test\"\n # disable limiter, otherwise problems will occure \n limiter: false\n public_instance: false\n\n# load the default json endpoint (set per default, just to make sure)\nsearch:\n formats:\n - html\n - json\n\n# DOI\u2011Resolver, for scientific papers\ndoi_resolvers:\n oadoi.org: 'https://oadoi.org/'\n doi.org: 'https://doi.org/'\ndefault_doi_resolver: 'oadoi.org'\n```\n\n2. in the searxng root terminal run docker compose up \n3. set the limit how many websites you want to search\n\n### 4. add parameters to the searching node:\n\n- click on **Add parameter** in the searxng node\n- Name: `categories`\n- Value: `news ` for news, `science` for research papers\n",
"height": 794,
"width": 670,
"color": 7
},
"id": "42104354-2343-454e-b48d-f4e25206b2a2",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
-580,
1420
]
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "c9aa5a9e-fffe-4cca-bd02-7c15b114f5d5",
"name": "results.url",
"value": "={{ $json.results.url }}",
"type": "string"
},
{
"id": "076427e6-c2cd-43f5-ba8b-3abc39fc491b",
"name": "query",
"value": "={{ $json.query }}",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
-220,
1180
],
"id": "e7967a97-c8cd-4b99-9734-78ba637e6968",
"name": "extract url and query"
},
{
"parameters": {
"method": "POST",
"url": "http://localhost:11235/crawl",
"sendBody": true,
"specifyBody": "json",
"jsonBody": "={\n \"urls\": [\"{{ $json.results.url }}\"],\n \"priority\": 10\n}\n",
"options": {}
},
"id": "0668123a-f6f7-411c-b5a5-7c74e72772be",
"name": "HTTP Request8",
"type": "n8n-nodes-base.httpRequest",
"typeVersion": 4.2,
"position": [
560,
1200
]
},
{
"parameters": {
"content": "## 1.3 Scrape the website with Crawl4ai\n\n1. Download the docker files: https://docs.crawl4ai.com/core/docker-deployment/\n2. follow the steps as shown in the documentation for local development and build and run locally with `docker compose up --build -d`\n\n\n",
"height": 194,
"width": 690,
"color": 7
},
"id": "99024475-d9a5-466b-929d-6b8dc0608152",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
240,
1420
]
},
{
"parameters": {
"content": "## 1.4 Store the websites in the Vector Database\n\n1. To store the data of the webiste in a vector database create a supabase account and create a projekt in the sql editor copy and run the following code, specified for embedding with ollama:\n\n```\n-- 1) pgvector-Extension (falls nicht bereits aktiv)\nCREATE EXTENSION IF NOT EXISTS vector;\n\n-- 2) Neue Tabelle mit 1024-dimensionalem Embedding\nCREATE TABLE documents (\n id bigserial PRIMARY KEY,\n content text,\n metadata jsonb,\n embedding vector(1024)\n);\n\n-- 3) IVFFlat-Index f\u00fcr schnelle \u00c4hnlichkeitssuche\nCREATE INDEX ON documents\nUSING ivfflat (embedding vector_l2_ops)\nWITH (lists = 100);\n\n-- 4) Funktion f\u00fcr 1024-dim-Vektoren anlegen\nCREATE OR REPLACE FUNCTION match_documents (\n query_embedding vector(1024),\n match_count int DEFAULT NULL,\n filter jsonb DEFAULT '{}'\n) RETURNS TABLE (\n id bigint,\n content text,\n metadata jsonb,\n similarity float\n)\nLANGUAGE plpgsql\nAS $$\nBEGIN\n RETURN QUERY\n SELECT\n docs.id AS id,\n docs.content AS content,\n docs.metadata AS metadata,\n 1 - (docs.embedding <=> query_embedding) AS similarity\n FROM documents AS docs\n WHERE docs.metadata @> filter\n ORDER BY docs.embedding <=> query_embedding\n LIMIT match_count;\nEND;\n$$;\n\n```\n\n2. Pull `ollama pull mxbai-embed-large:latest` and make sure the ollama server is already running with `ollana serve`\n\n",
"height": 954,
"width": 690,
"color": 7
},
"id": "1c5a9d2a-8c92-4034-a6c0-2f08d1a17dc0",
"name": "Sticky Note7",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
1720,
1200
]
},
{
"parameters": {
"promptType": "define",
"text": "=Basierend auf den vorliegenden Dokumenten und deinem Expertenwissen, beantworte diese Frage:\n\n{{ $json.frage }}",
"options": {
"systemPromptTemplate": "=Du bist Research-Assistant\u2122, Dr.-Level in E-Mobilit\u00e4t.\n\n# Regeln:\n\n1. Nutze ausschlie\u00dflich diesen Kontext:\n\n{context}\n\n2. Beantworte dann die Frage:\n\n{{ $json.frage }}\n\n3. verwende nur die relevanteste quelle und merke dir die url aus den metadaten, gib nur eine konkrete url zur\u00fcck! \n\n3. Antworte nur im JSON. Bef\u00fclle folgende Felder:\n\n- frage (string)\n- thema (string)\n- url (string)\n- zusammenfassung (string)\n- analyse (string)\n- ranking (string)\n\n"
}
},
"type": "@n8n/n8n-nodes-langchain.chainRetrievalQa",
"typeVersion": 1.5,
"position": [
220,
2760
],
"id": "d6204b6e-9dc4-434d-9f3d-71b005fe2385",
"name": "Question and Answer Chain",
"alwaysOutputData": true,
"retryOnFail": true,
"waitBetweenTries": 5000,
"onError": "continueRegularOutput"
},
{
"parameters": {
"topK": 20
},
"type": "@n8n/n8n-nodes-langchain.retrieverVectorStore",
"typeVersion": 1,
"position": [
340,
3040
],
"id": "397b624a-41a2-42ac-958d-709b2a22ec4f",
"name": "Vector Store Retriever"
},
{
"parameters": {
"tableName": {
"__rl": true,
"value": "documents",
"mode": "list",
"cachedResultName": "documents"
},
"options": {
"queryName": "match_documents"
}
},
"type": "@n8n/n8n-nodes-langchain.vectorStoreSupabase",
"typeVersion": 1.1,
"position": [
380,
3240
],
"id": "0735c115-b535-45f3-915c-8db9045d76a6",
"name": "Supabase Vector Store4",
"credentials": {
"supabaseApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"model": "mxbai-embed-large:latest"
},
"type": "@n8n/n8n-nodes-langchain.embeddingsOllama",
"typeVersion": 1,
"position": [
300,
3440
],
"id": "b203d676-4225-4401-bdc8-5eb591baac7c",
"name": "Embeddings Ollama3",
"credentials": {
"ollamaApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"modelName": "models/gemini-2.0-flash-001",
"options": {
"temperature": 0.2
}
},
"type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
"typeVersion": 1,
"position": [
200,
3040
],
"id": "ca1078d6-06de-457d-b5a5-acaa9815272e",
"name": "Google Gemini Chat Model5",
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"maxItems": 100
},
"type": "n8n-nodes-base.limit",
"typeVersion": 1,
"position": [
-40,
1180
],
"id": "2050262c-9fed-4565-b55d-80345b41dd72",
"name": "Limit"
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "00eb5335-583c-4756-aa39-6e79701b8495",
"name": "question",
"value": "Welche Trends, Technologien und Rahmenbedingungen sind aktuell und in den n\u00e4chsten Jahrzehnten relevant f\u00fcr die Entwicklung der Elektromobilit\u00e4t in Deutschland? Ber\u00fccksichtigen Sie dabei Infrastruktur (Lade-Netze, Stromversorgung), Politik und Regulierung (F\u00f6rderprogramme, Gesetzgebung), Wirtschaft (Marktentwicklung, Industriepartnerschaften), Technologie (Batterie-Innovation, Fahrzeugdesign) und gesellschaftliches Nutzungsverhalten.",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
-680,
2760
],
"id": "2a0be3f5-6bb7-4d2f-86d3-d93d59f9a9be",
"name": "Theme Question"
},
{
"parameters": {
"content": "# 1. Vectorstorage Workflow\n\n## Get and store data for topic: e-mobility\n\nThis is the workflow just to get the data and store it.\nAnalyzation will be done in the 2. workflow\n\n",
"height": 220,
"width": 490,
"color": 4
},
"id": "e17d2e2c-4be5-4db2-b585-07c76eab8abf",
"name": "Sticky Note8",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
-2240,
1100
]
},
{
"parameters": {
"content": "## 2.2 Set the analysing question and rank the best data\n\n- use **Theme Question** to edit the input question\n- Number of results by the Vector Store can be set in the **Vector Store Retreiver** Node\n",
"height": 214,
"width": 530,
"color": 7
},
"id": "aa973e9c-dffa-4f1f-81fd-f3b0b1d9dae3",
"name": "Sticky Note9",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
-720,
2340
]
},
{
"parameters": {
"rule": {
"interval": [
{
"triggerAtHour": 7
}
]
}
},
"type": "n8n-nodes-base.scheduleTrigger",
"typeVersion": 1.2,
"position": [
-1700,
1180
],
"id": "9e8c889b-019d-4e5d-9565-b7a297ad64e8",
"name": "Schedule Trigger"
},
{
"parameters": {
"content": "# 2. Analyzing workflow \n\n## Analyze the data\n\nThis is the workflow is necessary for the data analyzation. Simple with just one node\n\n\n",
"height": 220,
"width": 490,
"color": 5
},
"id": "2d7ba843-c34d-42b2-ac20-92287d380271",
"name": "Sticky Note10",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
-2220,
2340
]
},
{
"parameters": {
"content": "## Experimental: Analyze the data with an AI Agent \n\nThis is the workflow is necessary for the data analyzation, implemented with an AI Agent who calls a sub-workflow\n\n\n",
"height": 140,
"width": 490,
"color": 5
},
"id": "33b07640-987b-459c-8a86-ff7a48e8a9e9",
"name": "Sticky Note11",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
-2240,
5760
]
},
{
"parameters": {
"content": "## 3. Experimental: Sub-workflow for the AI Agent\n\nThis is the sub-workflow which gets the RAG Data, it needs to be copied in an extra workflow file outside of this one\n",
"height": 140,
"width": 490,
"color": 6
},
"id": "9320eedf-a03e-4050-a9a8-741da58dcaf0",
"name": "Sticky Note12",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
-780,
5760
]
},
{
"parameters": {
"promptType": "define",
"text": "=Basierend auf den vorliegenden Dokumenten und deinem Expertenwissen, beantworte diese Frage:\n\n{{ $json.question }}\n\nGib mir die Antwort als JSON-Array zur\u00fcck.\n",
"hasOutputParser": true,
"options": {
"systemMessage": "Du bist Research-Assistant\u2122, Dr.-Level in E-Mobilit\u00e4t. \nWenn Du Kontext brauchst, rufe **search_docs(query, 10)** auf. Du erh\u00e4ltst ein Array von 10 Dokumenten, jedes mit mindestens diesen Feldern:\n- url (string)\n- title (string)\n- snippet (string \u2013 kurzer Auszug)\n\nArbeite so:\n\n1. Rufe **search_docs({{ $json.parameters.query }}, 10)** auf (schreibe im Stream-of-Thought: \u201eI will call search_docs because\u2026\u201c). \n2. Ordne jedem der 10 Treffer einen eigenen Trend zu. \n3. F\u00fcr jedes Ergebnis baue ein JSON-Objekt mit diesen Feldern:\n - **uebertheme**: immer \u201eE-Mobilit\u00e4t\u201c \n - **theme**: kurzer Trend-Name (z. B. \u201ePlug and Charge\u201c) \n - **url**: exakte URL aus dem RAG-Treffer \n - **analysis**: 2\u20133 S\u00e4tze, die erkl\u00e4ren, **warum gerade dieses Dokument** den Trend untermauert (verweise auf Title oder Snippet) \n - **summary**: 1\u20132 S\u00e4tze pr\u00e4gnanter Zusammenfassung des Dokuments \n - **ranking**: Position 1\u201310 entsprechend Relevanz (1 = h\u00f6chst relevant) \n - **date**: heutiges Datum im Format YYYY-MM-DD\n\n4. Gib **nur** ein JSON-Array mit genau 10 solchen Objekten zur\u00fcck \u2013 sonst nichts.\n",
"maxIterations": 10
}
},
"type": "@n8n/n8n-nodes-langchain.agent",
"typeVersion": 1.9,
"position": [
-1220,
5760
],
"id": "03c1a13c-7c67-4273-91de-071f03b35f77",
"name": "AI Agent"
},
{
"parameters": {
"modelName": "models/gemini-2.0-flash-001",
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
"typeVersion": 1,
"position": [
-1280,
5980
],
"id": "1d474f31-6daa-419a-a5ba-12219cf6f5a7",
"name": "Google Gemini Chat Model",
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"rule": {
"interval": [
{}
]
}
},
"type": "n8n-nodes-base.scheduleTrigger",
"typeVersion": 1.2,
"position": [
-1660,
5760
],
"id": "83fa0089-6c65-4214-835d-96238efe447f",
"name": "Schedule Trigger2"
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "00eb5335-583c-4756-aa39-6e79701b8495",
"name": "question",
"value": " Welche Trends sind f\u00fcr E-Mobilit\u00e4t in Deutschland in den n\u00e4chsten 40 Jahren relevant?",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
-1440,
5760
],
"id": "108d6b05-dc6f-46de-bf2f-905db10c9ceb",
"name": "Theme Question2"
},
{
"parameters": {
"name": "search_docs",
"description": "search the vector data base to answer the question.",
"workflowId": {
"__rl": true,
"value": "0kjMhRTU8aA2NSAb",
"mode": "list",
"cachedResultName": "search_docs"
},
"workflowInputs": {
"mappingMode": "defineBelow",
"value": {
"name": "search_docs",
"description": "Search the vector database for semantically similar documents",
"parameters": "={\n \"query\": \"{{ $fromAI('query', '', 'string') }}\"\n}\n"
},
"matchingColumns": [],
"schema": [
{
"id": "name",
"displayName": "name",
"required": false,
"defaultMatch": false,
"display": true,
"canBeUsedToMatch": true,
"type": "string",
"removed": false
},
{
"id": "description",
"displayName": "description",
"required": false,
"defaultMatch": false,
"display": true,
"canBeUsedToMatch": true,
"type": "string",
"removed": false
},
{
"id": "parameters",
"displayName": "parameters",
"required": false,
"defaultMatch": false,
"display": true,
"canBeUsedToMatch": true,
"type": "object",
"removed": false
}
],
"attemptToConvertTypes": false,
"convertFieldsToString": false
}
},
"type": "@n8n/n8n-nodes-langchain.toolWorkflow",
"typeVersion": 2.1,
"position": [
-1120,
5980
],
"id": "6c752f83-a38d-436f-ab77-e8ff30cf8d70",
"name": "search_docs2"
},
{
"parameters": {
"jsonSchemaExample": "[\n {\n \"uebertheme\": \"E-Mobilit\u00e4t\",\n \"theme\": \"Plug and Charge\",\n \"url\": \"https://goingelectric.de/plug-and-charge-pilotprojekt\",\n \"analysis\": \"Dieses Pilotprojekt beschreibt im Detail, wie ISO 15118 die Ladeauthentifizierung automatisiert \u2013 besonders interessant sind die Feldtests in M\u00fcnchen und Berlin, die im Artikel erl\u00e4utert werden.\",\n \"summary\": \"Plug and Charge erm\u00f6glicht das automatische Starten und Abrechnen von Ladevorg\u00e4ngen ohne zus\u00e4tzliche Hardware oder App.\",\n \"ranking\": 1,\n \"date\": \"2025-04-25\"\n },\n {\n \"uebertheme\": \"E-Mobilit\u00e4t\",\n \"theme\": \"Bidirektionales Laden\",\n \"url\": \"https://bdew.de/v2g-studie-2024\",\n \"analysis\": \"Die BDEW-Studie zeigt anhand von Simulationen f\u00fcr das Berliner Netz, wie V2G bis zu 15 % Spitzenlast abfedern kann \u2013 interessant sind die Methodik und die konkreten Zahlen zur Netzstabilit\u00e4t.\",\n \"summary\": \"Bidirektionales Laden erlaubt es E-Autos, Strom ins Netz zur\u00fcckzuspeisen und so Netzschwankungen auszugleichen.\",\n \"ranking\": 2,\n \"date\": \"2025-04-25\"\n }\n]\n"
},
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"typeVersion": 1.2,
"position": [
-960,
5980
],
"id": "b8a276c7-0c85-4f78-82b1-4de976f815f9",
"name": "Structured Output Parser1"
},
{
"parameters": {
"promptType": "define",
"text": "Beantworte diese Frage mit den 10 besten Vektor-DB-Treffern und liefere eine wissenschaftlich fundierte Analyse:\n\nWelche Trends sind f\u00fcr E-Mobilit\u00e4t in Deutschland in den n\u00e4chsten 40 Jahren relevant?\n",
"options": {
"systemPromptTemplate": "=Du bist Research-Assistant\u2122, Dr.-Level in E-Mobilit\u00e4t.\nNutze ausschlie\u00dflich diesen Kontext:\n\n{context}\n\nBeantworte dann die Frage:\n\n{{ $json.parameters.query }}\n"
}
},
"type": "@n8n/n8n-nodes-langchain.chainRetrievalQa",
"typeVersion": 1.5,
"position": [
100,
5780
],
"id": "5afc0296-7521-4a78-8347-6d5c7da19189",
"name": "Question and Answer Chain1"
},
{
"parameters": {
"topK": 20
},
"type": "@n8n/n8n-nodes-langchain.retrieverVectorStore",
"typeVersion": 1,
"position": [
240,
6000
],
"id": "38f8e85d-a81b-41a5-8cdc-9e496d045dfc",
"name": "Vector Store Retriever1"
},
{
"parameters": {
"tableName": {
"__rl": true,
"value": "documents",
"mode": "list",
"cachedResultName": "documents"
},
"options": {
"queryName": "match_documents"
}
},
"type": "@n8n/n8n-nodes-langchain.vectorStoreSupabase",
"typeVersion": 1.1,
"position": [
280,
6200
],
"id": "6a6edecc-0b45-4b4b-ae61-5959683c11d3",
"name": "Supabase Vector Store5",
"credentials": {
"supabaseApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"model": "mxbai-embed-large:latest"
},
"type": "@n8n/n8n-nodes-langchain.embeddingsOllama",
"typeVersion": 1,
"position": [
200,
6400
],
"id": "d6101f48-ebbe-4d23-9439-6fa636382fbd",
"name": "Embeddings Ollama4",
"credentials": {
"ollamaApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"modelName": "models/gemini-2.0-flash-001",
"options": {
"temperature": 0.2
}
},
"type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
"typeVersion": 1,
"position": [
0,
6000
],
"id": "93668cb2-ef14-46d5-a1ae-0db16a273bf1",
"name": "Google Gemini Chat Model6",
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"inputSource": "jsonExample",
"jsonExample": "{\n \"name\": \"search_docs\",\n \"description\": \"Search the vector database for semantically similar documents\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"query\": {\n \"type\": \"string\",\n \"description\": \"The user\u2019s question\"\n }\n },\n \"required\": [\"query\"]\n }\n}\n"
},
"id": "0095816d-ab61-42ab-ad3b-7f622e84923c",
"typeVersion": 1.1,
"name": "search_docs1",
"type": "n8n-nodes-base.executeWorkflowTrigger",
"position": [
-200,
5780
]
},
{
"parameters": {
"promptType": "define",
"text": "=Basierend auf dieser Fragestellung:\n\n{{ $json.question }}\n\ngeneriere einen prompt f\u00fcr die Vectordatenbank um die relevantesten Dokumente herauszufinden, fasse dich vom unfang her so kurz wie die frage selbst, nutze unterschiedliche variationen damit immer eine neue query rauskommt.\n\nKein Einleitungstext, kein schnickschnack nur eine einzige Frage!",
"messages": {
"messageValues": [
{
"message": "=Du bist ein wissenschaftler und stellst dir die frage nach den zukunftstrends der e mobilit\u00e4t in deutschland. dein output ist eine einzige frage vom unfang her so gro\u00df wie die \u00fcbergebene frage: {{ $json.question }} varieere und sei kreativ in der formulierung, blicke \u00fcber den tellerand der urspr\u00fcnglichen frage hinaus!"
}
]
}
},
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"typeVersion": 1.6,
"position": [
-420,
2760
],
"id": "94b23fb5-41ed-431a-bf76-e7ef6ea0abd9",
"name": "Basic LLM Chain"
},
{
"parameters": {
"modelName": "models/gemini-2.0-flash-001",
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
"typeVersion": 1,
"position": [
-440,
3020
],
"id": "aac0344d-4b52-4a2d-b73f-86f64c0fae7e",
"name": "Google Gemini Chat Model1",
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "3f8d891d-fbb8-4688-bd75-591140b4bc82",
"name": "frage",
"value": "={{ $json.text }}",
"type": "string"
},
{
"id": "3f676690-e721-4aeb-a7cd-d439d878ee31",
"name": "thema",
"value": "",
"type": "string"
},
{
"id": "df01242b-0579-4b52-983d-b56aa1cca60c",
"name": "url",
"value": "",
"type": "string"
},
{
"id": "97533a06-06cd-44ef-a2d4-5d6c3c599fb4",
"name": "zusammenfassung",
"value": "",
"type": "string"
},
{
"id": "779dcfe4-3c40-482d-b48a-2df4fc13d4df",
"name": "analysen",
"value": "",
"type": "string"
},
{
"id": "338f90d9-89b6-4435-b2db-9991994e6af2",
"name": "ranking",
"value": 0,
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
-20,
2760
],
"id": "a13208bd-d991-4be0-8a76-07470dc0a6f0",
"name": "Edit Fields"
},
{
"parameters": {
"text": "={{ $json.response }}",
"attributes": {
"attributes": [
{
"name": "frage",
"description": "die urspr\u00fcngliche frage auf die das llm geantwortet hat",
"required": true
},
{
"name": "thema",
"description": "das thema kurz als \u00fcberschrift",
"required": true
},
{
"name": "url",
"description": "die angegebene url",
"required": true
},
{
"name": "zusammenfassung",
"description": "die zusammenfassung des textes",
"required": true
},
{
"name": "analysen",
"description": "die analyse des textes",
"required": true
},
{
"name": "ranking",
"description": "das ranking der relevanz",
"required": true
}
]
},
"options": {
"systemPromptTemplate": "You are an expert extraction algorithm.\nOnly extract relevant information from the text.\nIf you do not know the value of an attribute asked to extract, leave it blank"
}
},
"type": "@n8n/n8n-nodes-langchain.informationExtractor",
"typeVersion": 1,
"position": [
640,
2760
],
"id": "9f6d2845-4b0e-4298-be9a-c658c74294fc",
"name": "Information Extractor",
"alwaysOutputData": true,
"retryOnFail": true,
"waitBetweenTries": 5000,
"onError": "continueRegularOutput"
},
{
"parameters": {
"modelName": "models/gemini-2.0-flash-001",
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
"typeVersion": 1,
"position": [
700,
3040
],
"id": "be80dd0a-59f8-4e5e-818f-9c440d459433",
"name": "Google Gemini Chat Model4",
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "4ea899ff-ae1d-4dcb-8cd9-03cb799ac308",
"name": "output.frage",
"value": "={{ $json.output.frage }}",
"type": "string"
},
{
"id": "0fccd4e0-03e9-4874-bec9-8672ec86b722",
"name": "output.thema",
"value": "={{ $json.output.thema }}",
"type": "string"
},
{
"id": "08a4f0a2-1478-4e37-b4f8-1573e48787df",
"name": "output.url",
"value": "={{ $json.output.url }}",
"type": "string"
},
{
"id": "48d62b56-6363-4203-8fac-0965eec25dae",
"name": "output.zusammenfassung",
"value": "={{ $json.output.zusammenfassung }}",
"type": "string"
},
{
"id": "47f5ae98-881a-46ba-aac8-3017fe3a460b",
"name": "output.analysen",
"value": "={{ $json.output.analysen }}",
"type": "string"
},
{
"id": "ef909638-41a1-4776-a0c9-b432df090d0e",
"name": "output.ranking",
"value": "={{ $json.output.ranking }}",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
980,
2760
],
"id": "8e0098d3-0062-4d56-8e10-3792a3315278",
"name": "Edit Fields1"
},
{
"parameters": {
"options": {}
},
"type": "n8n-nodes-base.splitInBatches",
"typeVersion": 3,
"position": [
-960,
2580
],
"id": "0f2b5b94-46a5-4fdb-a1c5-f287e3eaef41",
"name": "Loop Over Items1"
},
{
"parameters": {
"jsCode": "// Anzahl der Wiederholungen\nconst count = 7;\n\n// Erzeuge ein Array mit 'count' leeren Objekten\nconst items = [];\nfor (let i = 0; i < count; i++) {\n items.push({ json: {} });\n}\n\n// Gib die Items zur\u00fcck\nreturn items;\n"
},
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
-1660,
2580
],
"id": "f308b605-d778-406a-a351-bf8ef29b3dad",
"name": "Code"
},
{
"parameters": {},
"type": "n8n-nodes-base.merge",
"typeVersion": 3.1,
"position": [
1220,
2580
],
"id": "c03a4b69-05ac-418b-a0c7-d91436c08962",
"name": "Merge"
},
{
"parameters": {
"promptType": "define",
"text": "=Du erh\u00e4ltst ein JSON-Array unter dem Feld {{ JSON.stringify($json.data) }}. \nJedes Element hat die Felder frage, thema, url, zusammenfassung, analyse, ranking. \nW\u00e4hle daraus die 2 **relevantesten** Items aus und liefere ein neues JSON-Array mit genau diesen beiden Objekten. \nF\u00fcge jedem Objekt das Feld \"begr\u00fcndung\" hinzu, in dem Du in 1\u20132 S\u00e4tzen erkl\u00e4rst, warum gerade diese Quelle so gew\u00e4hlt wurde.\n\n\nDu erh\u00e4ltst ein JSON-Array unter dem Feld {{ JSON.stringify($json.data) }}. \nJedes Element hat die Felder frage, thema, url, zusammenfassung, analysen, ranking.\n\nW\u00e4hle die 2 relevantesten Objekte aus, f\u00fcge jedem das Feld \"begr\u00fcndung\" hinzu, in dem Du in 1\u20132 S\u00e4tzen erkl\u00e4rst, warum gerade diese Quelle so gew\u00e4hlt wurde und gib **ausschlie\u00dflich** ein g\u00fcltiges JSON-Array zur\u00fcck.\n\n",
"hasOutputParser": true,
"messages": {
"messageValues": [
{
"message": "=Du bist eine Expertin f\u00fcr Elektrotechnik und Wirtschaft am Standort Deutschland und analysierst auf Professoren-Level f\u00fcr die Forschung.\n"
}
]
}
},
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"typeVersion": 1.6,
"position": [
1680,
2580
],
"id": "2319b89d-7b01-41fe-94bc-786a5d66d2fe",
"name": "Basic LLM Chain1",
"retryOnFail": true,
"waitBetweenTries": 5000,
"alwaysOutputData": true,
"onError": "continueRegularOutput"
},
{
"parameters": {
"modelName": "models/gemini-2.0-flash-001",
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
"typeVersion": 1,
"position": [
1600,
2780
],
"id": "9ad69e58-1678-4bf5-9f69-f804a54531e7",
"name": "Google Gemini Chat Model7",
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"aggregate": "aggregateAllItemData",
"options": {}
},
"type": "n8n-nodes-base.aggregate",
"typeVersion": 1,
"position": [
1440,
2580
],
"id": "ff350a1e-b24f-463a-9297-0388645aaa04",
"name": "Aggregate"
},
{
"parameters": {
"schemaType": "manual",
"inputSchema": "{\n \"type\": \"array\",\n \"minItems\": 2,\n \"maxItems\": 2,\n \"items\": {\n \"type\": \"object\",\n \"properties\": {\n \"frage\": { \"type\": \"string\" },\n \"thema\": { \"type\": \"string\" },\n \"url\": { \"type\": \"string\", \"format\": \"uri\" },\n \"zusammenfassung\": { \"type\": \"string\" },\n \"analysen\": { \"type\": \"string\" },\n \"ranking\": { \"type\": \"string\" },\n \"begr\u00fcndung\": { \"type\": \"string\" }\n },\n \"required\": [\"frage\",\"thema\",\"url\",\"zusammenfassung\",\"analyse\",\"ranking\",\"begr\u00fcndung\"]\n }\n}\n"
},
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"typeVersion": 1.2,
"position": [
1800,
2780
],
"id": "dd4b55b7-1d0f-472c-a80d-1d8ce835928b",
"name": "Structured Output Parser2"
},
{
"parameters": {
"tableId": "quellen",
"fieldsUi": {
"fieldValues": [
{
"fieldId": "frage",
"fieldValue": "={{ $json.output[0].frage }}"
},
{
"fieldId": "thema",
"fieldValue": "={{ $json.output[0].thema }}"
},
{
"fieldId": "url",
"fieldValue": "={{ $json.output[0].url }}"
},
{
"fieldId": "zusammenfassung",
"fieldValue": "={{ $json.output[0].zusammenfassung }}"
},
{
"fieldId": "analysen",
"fieldValue": "={{ $json.output[0].analysen }}"
},
{
"fieldId": "ranking",
"fieldValue": "={{ $json.output[0].ranking }}"
},
{
"fieldId": "begr\u00fcndung",
"fieldValue": "={{ $json.output[0]['begr\u00fcndung'] }}"
}
]
}
},
"type": "n8n-nodes-base.supabase",
"typeVersion": 1,
"position": [
2340,
2740
],
"id": "3156807a-574d-4df2-a204-3ee0d49d2911",
"name": "Supabase",
"credentials": {
"supabaseApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"options": {}
},
"type": "n8n-nodes-base.splitInBatches",
"typeVersion": 3,
"position": [
2140,
2580
],
"id": "f4dc2821-82d4-4a48-b055-edc1c7199956",
"name": "Loop Over Items2"
},
{
"parameters": {
"content": "## 2.3 Let the Model analyze the best results\n\n- in this node the model analyzes the results and outputs an array only with the two best results\n- the amount of results the model gives back can be defined in the prompt\n",
"height": 194,
"width": 530,
"color": 7
},
"id": "83113789-8ebf-4b0d-933e-2b1ab0cc47a9",
"name": "Sticky Note13",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
1220,
2340
]
},
{
"parameters": {
"content": "## 2.4 Store the best results in the database\n\n- it loops over the best results and stores them in the database\n- Setup the database with the following sql query in the sql editor in supabase:\n\n```\ncreate extension if not exists \"pgcrypto\";\n\n\ncreate table public.quellen (\n id uuid primary key default gen_random_uuid(),\n frage text not null,\n thema text not null,\n url text not null,\n zusammenfassung text, -- <\u2014 Komma hier!\n analysen text, -- jetzt klappt die n\u00e4chste Zeile\n ranking text,\n \"begr\u00fcndung\" text, -- Umlaut-Spalte in Anf\u00fchrungszeichen\n created_at timestamp with time zone default now()\n);\n```\n",
"height": 434,
"width": 810,
"color": 7
},
"id": "46481bbf-95f9-4c2c-8748-481e675c4c7a",
"name": "Sticky Note14",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
2560,
2560
]
},
{
"parameters": {
"content": "## 2.1 Setting the amount of retrieved items from the vector store\n\n- set the amount of retrieved items with `const count`\n- the amount will define how many resources the AI will analyze\n",
"height": 214,
"width": 530,
"color": 7
},
"id": "73765dce-6967-45e2-96cd-e0fdf47a85e3",
"name": "Sticky Note15",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
-1660,
2340
]
},
{
"parameters": {
"content": "## 2.3 Let the LLM retreive Data from the Vector store\n\n- the LLM will retreive the data from supabase\n- connect supbase, gemini and ollama credentials\n",
"height": 214,
"width": 530,
"color": 7
},
"id": "fe6418c6-8604-408f-8372-80f2c5a43d96",
"name": "Sticky Note16",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
200,
2340
]
},
{
"parameters": {},
"type": "n8n-nodes-base.noOp",
"name": "Replace Me",
"typeVersion": 1,
"position": [
-1660,
4040
],
"id": "2abd2359-c451-42ad-9829-540cb079be8a"
},
{
"parameters": {
"content": "# 3. Continue the Workflow\n\n- to continue the workflow replace this node with the workflow 1 and 2 by copy and pasting them.\n- this way it is possible to analyze multiple different topics in one continious workflow\n- subject topics need to be changed in the `set search request` node and the `theme topic` node\n\n",
"height": 220,
"width": 490,
"color": 7
},
"id": "e8351e79-264b-44ef-82fe-c504b58f52fd",
"name": "Sticky Note17",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
-2220,
4000
]
}
],
"connections": {
"Loop Over Items": {
"main": [
[
{
"node": "Code",
"type": "main",
"index": 0
}
],
[
{
"node": "HTTP Request8",
"type": "main",
"index": 0
}
]
]
},
"Default Data Loader": {
"ai_document": [
[
{
"node": "Supabase Vector Store",
"type": "ai_document",
"index": 0
}
]
]
},
"Character Text Splitter": {
"ai_textSplitter": [
[
{
"node": "Default Data Loader",
"type": "ai_textSplitter",
"index": 0
}
]
]
},
"Supabase Vector Store": {
"main": [
[
{
"node": "Loop Over Items",
"type": "main",
"index": 0
}
]
]
},
"Split Out1": {
"main": [
[
{
"node": "Supabase Vector Store",
"type": "main",
"index": 0
}
]
]
},
"Embeddings Ollama": {
"ai_embedding": [
[
{
"node": "Supabase Vector Store",
"type": "ai_embedding",
"index": 0
}
]
]
},
"Structured Output Parser": {
"ai_outputParser": [
[
{
"node": "Get different search variations",
"type": "ai_outputParser",
"index": 0
}
]
]
},
"searxng": {
"main": [
[
{
"node": "Split Out3",
"type": "main",
"index": 0
}
]
]
},
"Google Gemini Chat Model2": {
"ai_languageModel": [
[
{
"node": "Get different search variations",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Split Out2": {
"main": [
[
{
"node": "searxng",
"type": "main",
"index": 0
}
]
]
},
"Split Out3": {
"main": [
[
{
"node": "extract url and query",
"type": "main",
"index": 0
}
]
]
},
"Set search request": {
"main": [
[
{
"node": "Get different search variations",
"type": "main",
"index": 0
}
]
]
},
"Get different search variations": {
"main": [
[
{
"node": "Split Out2",
"type": "main",
"index": 0
}
]
]
},
"extract url and query": {
"main": [
[
{
"node": "Limit",
"type": "main",
"index": 0
}
]
]
},
"HTTP Request8": {
"main": [
[
{
"node": "Split Out1",
"type": "main",
"index": 0
}
]
]
},
"Vector Store Retriever": {
"ai_retriever": [
[
{
"node": "Question and Answer Chain",
"type": "ai_retriever",
"index": 0
}
]
]
},
"Supabase Vector Store4": {
"ai_vectorStore": [
[
{
"node": "Vector Store Retriever",
"type": "ai_vectorStore",
"index": 0
}
]
]
},
"Embeddings Ollama3": {
"ai_embedding": [
[
{
"node": "Supabase Vector Store4",
"type": "ai_embedding",
"index": 0
}
]
]
},
"Google Gemini Chat Model5": {
"ai_languageModel": [
[
{
"node": "Question and Answer Chain",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Limit": {
"main": [
[
{
"node": "Loop Over Items",
"type": "main",
"index": 0
}
]
]
},
"Question and Answer Chain": {
"main": [
[
{
"node": "Information Extractor",
"type": "main",
"index": 0
}
]
]
},
"Theme Question": {
"main": [
[
{
"node": "Basic LLM Chain",
"type": "main",
"index": 0
}
]
]
},
"Schedule Trigger": {
"main": [
[
{
"node": "Set search request",
"type": "main",
"index": 0
}
]
]
},
"Google Gemini Chat Model": {
"ai_languageModel": [
[
{
"node": "AI Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Schedule Trigger2": {
"main": [
[
{
"node": "Theme Question2",
"type": "main",
"index": 0
}
]
]
},
"Theme Question2": {
"main": [
[
{
"node": "AI Agent",
"type": "main",
"index": 0
}
]
]
},
"search_docs2": {
"ai_tool": [
[
{
"node": "AI Agent",
"type": "ai_tool",
"index": 0
}
]
]
},
"Structured Output Parser1": {
"ai_outputParser": [
[
{
"node": "AI Agent",
"type": "ai_outputParser",
"index": 0
}
]
]
},
"Vector Store Retriever1": {
"ai_retriever": [
[
{
"node": "Question and Answer Chain1",
"type": "ai_retriever",
"index": 0
}
]
]
},
"Supabase Vector Store5": {
"ai_vectorStore": [
[
{
"node": "Vector Store Retriever1",
"type": "ai_vectorStore",
"index": 0
}
]
]
},
"Embeddings Ollama4": {
"ai_embedding": [
[
{
"node": "Supabase Vector Store5",
"type": "ai_embedding",
"index": 0
}
]
]
},
"Google Gemini Chat Model6": {
"ai_languageModel": [
[
{
"node": "Question and Answer Chain1",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"search_docs1": {
"main": [
[
{
"node": "Question and Answer Chain1",
"type": "main",
"index": 0
}
]
]
},
"Google Gemini Chat Model1": {
"ai_languageModel": [
[
{
"node": "Basic LLM Chain",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Basic LLM Chain": {
"main": [
[
{
"node": "Edit Fields",
"type": "main",
"index": 0
}
]
]
},
"Edit Fields": {
"main": [
[
{
"node": "Question and Answer Chain",
"type": "main",
"index": 0
}
]
]
},
"Google Gemini Chat Model4": {
"ai_languageModel": [
[
{
"node": "Information Extractor",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Information Extractor": {
"main": [
[
{
"node": "Edit Fields1",
"type": "main",
"index": 0
}
]
]
},
"Loop Over Items1": {
"main": [
[
{
"node": "Merge",
"type": "main",
"index": 0
}
],
[
{
"node": "Theme Question",
"type": "main",
"index": 0
}
]
]
},
"Edit Fields1": {
"main": [
[
{
"node": "Loop Over Items1",
"type": "main",
"index": 0
}
]
]
},
"Code": {
"main": [
[
{
"node": "Loop Over Items1",
"type": "main",
"index": 0
}
]
]
},
"Merge": {
"main": [
[
{
"node": "Aggregate",
"type": "main",
"index": 0
}
]
]
},
"Google Gemini Chat Model7": {
"ai_languageModel": [
[
{
"node": "Basic LLM Chain1",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Aggregate": {
"main": [
[
{
"node": "Basic LLM Chain1",
"type": "main",
"index": 0
}
]
]
},
"Basic LLM Chain1": {
"main": [
[
{
"node": "Loop Over Items2",
"type": "main",
"index": 0
}
]
]
},
"Structured Output Parser2": {
"ai_outputParser": [
[
{
"node": "Basic LLM Chain1",
"type": "ai_outputParser",
"index": 0
}
]
]
},
"Loop Over Items2": {
"main": [
[
{
"node": "Replace Me",
"type": "main",
"index": 0
}
],
[
{
"node": "Supabase",
"type": "main",
"index": 0
}
]
]
},
"Supabase": {
"main": [
[
{
"node": "Loop Over Items2",
"type": "main",
"index": 0
}
]
]
},
"Replace Me": {
"main": [
[]
]
}
},
"active": false,
"settings": {
"executionOrder": "v1"
},
"versionId": "dd41bbd4-e79f-4aab-af99-6c1fde224d75",
"meta": {
"templateCredsSetupCompleted": true
},
"id": "s1ACKPJ7YkhwiIiv",
"tags": []
}
Credentials you'll need
Each integration node will prompt for credentials when you import. We strip credential IDs before publishing — you'll add your own.
googlePalmApiollamaApisupabaseApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
crawl4 ai. Uses documentDefaultDataLoader, textSplitterCharacterTextSplitter, vectorStoreSupabase, embeddingsOllama. Scheduled trigger; 65 nodes.
Source: https://github.com/kevinboehmisch/n8n-automation/blob/882d1601e729ddced8fc5f0cbcd16410325e92e7/n8n-template/search_workflow_locally.json — original creator credit. Request a take-down →
Related workflows
Workflows that share integrations, category, or trigger type with this one. All free to copy and import.
This template is a complete, hands-on tutorial for building a RAG (Retrieval-Augmented Generation) pipeline. In simple terms, you'll teach an AI to become an expert on a specific topic—in this case, t
Memory Buffer Window, Supabase Vector Store, Document Default Data Loader +8
Supercharge your trading decisions with this end-to-end AI automation that connects market intelligence, technical analysis, and automated trade execution — all without manual intervention.
Tool Think, Supabase Vector Store, OpenAI Embeddings +14
Camila IA. Uses postgres, crypto, redis, agent. Webhook trigger; 92 nodes.
Postgres, Crypto, Redis +13
This comprehensive workflow bundle is designed as a powerful starter kit, enabling you to build a multi-functional AI assistant on Telegram. It seamlessly integrates AI-powered voice interactions, an
Telegram Trigger, Telegram, OpenAI +19
Hi! I’m Amanda, a creator of intelligent automations using n8n and Make. I’ve been building AI-powered workflows for over 2 years, always focused on usability and innovation. This one here is very spe
OpenAI Chat, Redis, OpenAI +11