Yusuke Yamamoto 
This workflow corresponds to n8n.io template #9558 — we link there as the canonical source.
This workflow follows the Agent → OpenRouter Chat 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 →
{
"id": "t0pM31b82LN8h8EX",
"meta": {
"templateCredsSetupCompleted": true
},
"name": "AI Recipe Generator for Telegram (from Image or Text)",
"tags": [],
"nodes": [
{
"id": "if_node_001",
"name": "Check Input Type",
"type": "n8n-nodes-base.if",
"position": [
400,
-16
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 1,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "condition_001",
"operator": {
"type": "string",
"operation": "notExists",
"singleValue": true
},
"leftValue": "={{ $json.message.text }}",
"rightValue": ""
}
]
}
},
"typeVersion": 2
},
{
"id": "ai_agent_vision_001",
"name": "AI Vision Agent",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
704,
-128
],
"parameters": {
"text": "=Please analyze this food image and identify all visible ingredients. List them clearly.\n\nImage URL: {{ $json.body.events[0].message.id }}",
"options": {
"systemMessage": "=You are a food ingredient identification expert. Analyze food images and identify all visible ingredients accurately. Return a structured list of ingredients."
},
"promptType": "define",
"hasOutputParser": true
},
"typeVersion": 2
},
{
"id": "openai_vision_model_001",
"name": "OpenAI Vision Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenRouter",
"position": [
672,
16
],
"parameters": {
"model": "openai/gpt-4-vision-preview",
"options": {
"temperature": 0.3
}
},
"credentials": {
"openRouterApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "output_parser_001",
"name": "Ingredient Parser",
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"position": [
896,
16
],
"parameters": {
"schemaType": "manual",
"inputSchema": "{\n \"type\": \"object\",\n \"properties\": {\n \"ingredients\": {\n \"type\": \"array\",\n \"items\": {\n \"type\": \"string\"\n },\n \"description\": \"List of identified ingredients\"\n }\n },\n \"required\": [\"ingredients\"]\n}"
},
"typeVersion": 1.2
},
{
"id": "set_ingredients_001",
"name": "Format Ingredients",
"type": "n8n-nodes-base.set",
"position": [
1088,
-128
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "assignment_001",
"name": "ingredients_text",
"type": "string",
"value": "={{ $json.output.ingredients.join(', ') }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "set_text_ingredients_001",
"name": "Get Text Ingredients",
"type": "n8n-nodes-base.set",
"position": [
688,
144
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "assignment_002",
"name": "ingredients_text",
"type": "string",
"value": "={{ $json.message.text }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "recipe_generator_001",
"name": "Recipe Generator",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
1264,
144
],
"parameters": {
"text": "=Based on these ingredients: {{ $json.ingredients_text }}\n\nPlease suggest 3 delicious recipes that can be made with these ingredients. Consider both Japanese and international cuisine options. Text is Japanese.",
"options": {
"systemMessage": "You are a professional chef and recipe expert. Create detailed, easy-to-follow recipes based on available ingredients. Include both traditional Japanese dishes and international cuisine. Always provide cooking time, difficulty level, and step-by-step instructions."
},
"promptType": "define",
"hasOutputParser": true
},
"typeVersion": 2
},
{
"id": "openai_recipe_model_001",
"name": "OpenAI Recipe Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenRouter",
"position": [
1264,
320
],
"parameters": {
"model": "openai/gpt-4o-mini",
"options": {
"temperature": 0.7
}
},
"credentials": {
"openRouterApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "recipe_parser_001",
"name": "Recipe Parser",
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"position": [
1456,
320
],
"parameters": {
"schemaType": "manual",
"inputSchema": "{\n \"type\": \"object\",\n \"properties\": {\n \"recipes\": {\n \"type\": \"array\",\n \"items\": {\n \"type\": \"object\",\n \"properties\": {\n \"name\": {\n \"type\": \"string\",\n \"description\": \"Recipe name\"\n },\n \"cuisine\": {\n \"type\": \"string\",\n \"description\": \"Cuisine type (e.g., Japanese, Italian)\"\n },\n \"difficulty\": {\n \"type\": \"string\",\n \"description\": \"Difficulty level (Easy, Medium, Hard)\"\n },\n \"cookingTime\": {\n \"type\": \"string\",\n \"description\": \"Total cooking time\"\n },\n \"servings\": {\n \"type\": \"number\",\n \"description\": \"Number of servings\"\n },\n \"ingredients\": {\n \"type\": \"array\",\n \"items\": {\n \"type\": \"string\"\n },\n \"description\": \"Full ingredient list with quantities\"\n },\n \"instructions\": {\n \"type\": \"array\",\n \"items\": {\n \"type\": \"string\"\n },\n \"description\": \"Step-by-step cooking instructions\"\n }\n },\n \"required\": [\"name\", \"cuisine\", \"difficulty\", \"cookingTime\", \"servings\", \"ingredients\", \"instructions\"]\n },\n \"description\": \"Array of recipe suggestions\"\n }\n },\n \"required\": [\"recipes\"]\n}"
},
"typeVersion": 1.2
},
{
"id": "format_response_001",
"name": "Format Response",
"type": "n8n-nodes-base.set",
"position": [
1680,
144
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "response_004",
"name": "message",
"type": "string",
"value": "={{\n(() => {\n return $json.output.recipes.map((r, index) => {\n return (\n `\ud83d\udcd6 \u30ec\u30b7\u30d4${index + 1}\uff1a${r.name}\\n` +\n `\ud83c\udf10 \u30b8\u30e3\u30f3\u30eb\uff1a${r.cuisine}\\n` +\n `\u23f1 \u8abf\u7406\u6642\u9593\uff1a${r.cookingTime}\\n` +\n `\ud83c\udf9a \u96e3\u6613\u5ea6\uff1a${r.difficulty}\\n` +\n `\ud83d\udc65 \u76ee\u5b89\u4eba\u6570\uff1a${r.servings}\u4eba\u5206\\n\\n` +\n `\ud83d\uded2 \u6750\u6599:\\n${r.ingredients.map(i => `- ${i}`).join(\"\\n\")}\\n\\n` +\n `\ud83d\udc63 \u624b\u9806:\\n${r.instructions.map((step, i) => `${i + 1}. ${step}`).join(\"\\n\")}`\n );\n }).join(\"\\n\\n------------------------\\n\\n\");\n})()\n}}\n"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "684b387e-b22d-47a2-8db3-5367850fa33d",
"name": "Telegram Trigger",
"type": "n8n-nodes-base.telegramTrigger",
"position": [
176,
-16
],
"parameters": {
"updates": [
"message"
],
"additionalFields": {
"download": true
}
},
"credentials": {
"telegramApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.2
},
{
"id": "812b9b19-c320-4a6b-a28f-7bf80a7863bd",
"name": "Send a text message",
"type": "n8n-nodes-base.telegram",
"position": [
1872,
144
],
"parameters": {
"text": "={{ $json.message }}",
"chatId": "={{ $('Telegram Trigger').item.json.message.chat.id }}",
"additionalFields": {
"appendAttribution": false
}
},
"credentials": {
"telegramApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.2
},
{
"id": "a9e0f1b2-c3d4-e5f6-a7b8-c9d0e1f2a3b4",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
160,
-320
],
"parameters": {
"color": 6,
"width": 500,
"height": 260,
"content": "### Phase 1: User Input via Telegram\n\n- **Telegram Trigger**: Starts the workflow when a user sends a message or image.\n\n- **Check Input Type**: This crucial IF node checks if the message contains text. If not, it assumes an image was sent and routes the flow to the Vision AI path. This allows the workflow to handle both input types."
},
"typeVersion": 1
},
{
"id": "b1c2d3e4-f5a6-b7c8-d9e0-f1a2b3c4d5e6",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
624,
-448
],
"parameters": {
"color": 3,
"width": 500,
"height": 380,
"content": "### Phase 2: Ingredient Extraction (Dual Path)\nThis section's goal is to create a standardized list of ingredients for the next phase.\n\n- **Image Path (Top)**: The `AI Vision Agent` analyzes the photo to identify ingredients. The `Ingredient Parser` then structures this into a clean list, and `Format Ingredients` converts it to a comma-separated string.\n\n- **Text Path (Bottom)**: The `Get Text Ingredients` node simply takes the user's text directly."
},
"typeVersion": 1
},
{
"id": "c3d4e5f6-a7b8-c9d0-e1f2-a3b4c5d6e7f8",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
1216,
-192
],
"parameters": {
"color": 5,
"width": 500,
"height": 360,
"content": "### Phase 3: AI Recipe Generation\n\n- **Recipe Generator**: This is the main creative AI. It takes the standardized ingredient list (from either path) and acts as a professional chef to generate 3 detailed recipes.\n\n- **Recipe Parser**: This is critical for reliability. It forces the AI's creative output into a structured JSON format (name, difficulty, instructions, etc.). This ensures the data is clean and predictable for the final formatting step."
},
"typeVersion": 1
},
{
"id": "d5e6f7a8-b9c0-d1e2-f3a4-b5c6d7e8f9a0",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
1648,
-128
],
"parameters": {
"color": 4,
"width": 500,
"height": 280,
"content": "### Phase 4: Format & Send Response\n\n- **Format Response**: This node uses a JavaScript expression to transform the structured recipe JSON from the parser into a beautiful, emoji-rich, and easy-to-read message for the user.\n\n- **Send a text message**: The final, formatted text is sent back to the user on Telegram."
},
"typeVersion": 1
}
],
"active": false,
"settings": {
"timezone": "Asia/Tokyo",
"errorWorkflow": "",
"executionOrder": "v1",
"saveManualExecutions": true,
"saveExecutionProgress": true,
"saveDataErrorExecution": "all",
"saveDataSuccessExecution": "all"
},
"versionId": "30c406be-b5d8-4ba8-873f-8ae990fdbcb9",
"connections": {
"Recipe Parser": {
"ai_outputParser": [
[
{
"node": "Recipe Generator",
"type": "ai_outputParser",
"index": 0
}
]
]
},
"AI Vision Agent": {
"main": [
[
{
"node": "Format Ingredients",
"type": "main",
"index": 0
}
]
]
},
"Format Response": {
"main": [
[
{
"node": "Send a text message",
"type": "main",
"index": 0
}
]
]
},
"Check Input Type": {
"main": [
[
{
"node": "AI Vision Agent",
"type": "main",
"index": 0
}
],
[
{
"node": "Get Text Ingredients",
"type": "main",
"index": 0
}
]
]
},
"Recipe Generator": {
"main": [
[
{
"node": "Format Response",
"type": "main",
"index": 0
}
]
]
},
"Telegram Trigger": {
"main": [
[
{
"node": "Check Input Type",
"type": "main",
"index": 0
}
]
]
},
"Ingredient Parser": {
"ai_outputParser": [
[
{
"node": "AI Vision Agent",
"type": "ai_outputParser",
"index": 0
}
]
]
},
"Format Ingredients": {
"main": [
[
{
"node": "Recipe Generator",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Recipe Model": {
"ai_languageModel": [
[
{
"node": "Recipe Generator",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"OpenAI Vision Model": {
"ai_languageModel": [
[
{
"node": "AI Vision Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Get Text Ingredients": {
"main": [
[
{
"node": "Recipe Generator",
"type": "main",
"index": 0
}
]
]
}
}
}
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.
openRouterApitelegramApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
This n8n template demonstrates a multi-modal AI recipe assistant that suggests delicious recipes based on user input, delivered via Telegram. The workflow can uniquely handle two types of input: a photo of your ingredients or a simple text list.
Source: https://n8n.io/workflows/9558/ — 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.
RAG CHATBOT Main. Uses telegram, telegramTrigger, lmChatOpenAi, n8n-nodes-mcp. Event-driven trigger; 87 nodes.
Telegram, Telegram Trigger, OpenAI Chat +8
This workflow transforms your Telegram bot into an intelligent creative assistant. It can chat conversationally, fetch trending image prompts from PromptHero for inspiration, or perform a deep "remix"
Telegram Trigger, Output Parser Structured, Telegram +6
This workflow contains community nodes that are only compatible with the self-hosted version of n8n.
Output Parser Structured, Telegram, N8N Nodes Tesseractjs +14
This workflow acts as an AI-powered "Viral Architect" for YouTube creators. Simply send a video topic (e.g., "Kling 2.6") to your Telegram bot, and it will scrape top-performing competitor thumbnails,
Output Parser Structured, N8N Nodes Browseract, OpenRouter Chat +6
This workflow acts as an intelligent content engine. Simply send a link to your Telegram bot (e.g., a product page or news article), and it will automatically scrape the content, rewrite it into a hig
Google Gemini, Agent, OpenRouter Chat +7