{ "edges": [ { "to": "e404aef6-0642-425e-88ec-0b96f14d8cb5", "from": "30c1481e-453a-4ba1-94b8-3e227ae87ff9", "toPort": "0", "fromPort": "0" }, { "to": "5f451ebc-a5db-4e2c-b398-4e00b3338ebf", "from": "8a2fbb6e-d808-484d-8830-fbdd9fe633ba", "toPort": "0", "fromPort": "1" }, { "to": "0771a2c0-c6fa-425c-b896-0d516ea9b01b", "from": "5f451ebc-a5db-4e2c-b398-4e00b3338ebf", "toPort": "0", "fromPort": "0" }, { "to": "8a2fbb6e-d808-484d-8830-fbdd9fe633ba", "from": "ccaabf97-84a1-4fef-9ebb-5767ec7ee0a4", "toPort": "0", "fromPort": "0" }, { "to": "791c1603-2c0a-49e9-a910-f63f66e77d56", "from": "463fae5f-aba1-4acf-90a2-895976bc44a8", "toPort": "0", "fromPort": "ai_languageModel" }, { "to": "a00daa5d-259c-4416-b864-257e44ef5573", "from": "463fae5f-aba1-4acf-90a2-895976bc44a8", "toPort": "0", "fromPort": "ai_languageModel" }, { "to": "791c1603-2c0a-49e9-a910-f63f66e77d56", "from": "a00daa5d-259c-4416-b864-257e44ef5573", "toPort": "0", "fromPort": "ai_outputParser" }, { "to": "8a2fbb6e-d808-484d-8830-fbdd9fe633ba", "from": "0771a2c0-c6fa-425c-b896-0d516ea9b01b", "toPort": "0", "fromPort": "0" }, { "to": "791c1603-2c0a-49e9-a910-f63f66e77d56", "from": "e404aef6-0642-425e-88ec-0b96f14d8cb5", "toPort": "0", "fromPort": "0" }, { "to": "ccaabf97-84a1-4fef-9ebb-5767ec7ee0a4", "from": "791c1603-2c0a-49e9-a910-f63f66e77d56", "toPort": "0", "fromPort": "0" } ], "nodes": [ { "id": "463fae5f-aba1-4acf-90a2-895976bc44a8", "name": "Google Gemini Chat Model", "type": "llm", "params": { "options": {}, "provider": "google", "modelName": "models/gemini-2.5-pro" }, "position": [ 1296, 528 ] }, { "id": "a00daa5d-259c-4416-b864-257e44ef5573", "name": "Structured Output Parser", "type": "passthrough", "params": { "mode": "noOp", "autoFix": true, "jsonSchemaExample": "{\n\"Telegram\" : [\n \"🎧 Podcast Title 1\\n🎙 Show Name\\n\\nSummary text here...\\n\\n──────────────\\n\\n🎧 Podcast Title 2... [Message content up to 3500 chars] ... \\n\\n(To be continued...)\",\n \"(...Continued)\\n\\n🎧 Podcast Title 5\\n🎙 Show Name\\n\\nSummary text...\"\n]\n}" }, "position": [ 1440, 528 ] }, { "id": "8a2fbb6e-d808-484d-8830-fbdd9fe633ba", "name": "Loop Over Items", "type": "passthrough", "params": { "mode": "noOp", "options": {} }, "position": [ 2064, 320 ] }, { "id": "30c1481e-453a-4ba1-94b8-3e227ae87ff9", "name": "Schedule Daily", "type": "trigger", "params": { "kind": "n8n-nodes-base.scheduleTrigger", "rule": { "interval": [ {} ] } }, "position": [ 656, 320 ] }, { "id": "e404aef6-0642-425e-88ec-0b96f14d8cb5", "name": "Run Top Chart Podcast workflow", "type": "integration", "params": { "type": "WORKFLOW", "service": "browseract", "operation": "run", "workflowId": "72057695044160531", "__originalType": "n8n-nodes-browseract.browserAct", "workflowConfig": { "value": {}, "schema": [ { "id": "input-Apple_Podcast", "type": "string", "display": true, "removed": true, "required": false, "description": "If left blank, the default value defined in BrowserAct will be used.", "displayName": "Apple_Podcast", "defaultMatch": true } ], "mappingMode": "defineBelow", "matchingColumns": [ "input-Apple_Podcast" ], "attemptToConvertTypes": false, "convertFieldsToString": false }, "open_incognito_mode": false }, "position": [ 864, 320 ] }, { "id": "ccaabf97-84a1-4fef-9ebb-5767ec7ee0a4", "name": "Split Generated Items", "type": "passthrough", "params": { "mode": "splitOut", "field": "output.Telegram", "options": {}, "fieldToSplitOut": "output.Telegram" }, "position": [ 1856, 320 ] }, { "id": "5f451ebc-a5db-4e2c-b398-4e00b3338ebf", "name": "Avoid Rate Limits", "type": "passthrough", "params": { "mode": "noOp" }, "position": [ 2304, 336 ] }, { "id": "0771a2c0-c6fa-425c-b896-0d516ea9b01b", "name": "Send Podcast List to User", "type": "integration", "params": { "text": "={{ $('Loop Over Items').item.json[\"output.Telegram\"] }}", "chatId": "parameters.chatId==@Channel_ID (Use Channel ID or Chat ID )", "service": "telegram", "operation": "read", "additionalFields": { "parse_mode": "HTML" } }, "position": [ 2512, 336 ] }, { "id": "791c1603-2c0a-49e9-a910-f63f66e77d56", "name": "Analyze Podcast & Generate Post", "type": "llm", "params": { "text": "={{ $json.output.string }}", "prompt": "={{ $json.output.string }}", "options": { "systemMessage": "You are a Telegram Content Formatter and Publisher. Your task is to take a raw JSON input containing podcast or news items and convert it into a structured JSON array of strings, formatted specifically for Telegram's HTML parsing mode.\n\n\nProcessing Rules:\n\nContent Formatting (Per Item): For each item in the input list, generate a text block using the following HTML tags supported by Telegram (, , ):\n\nHeadline: Use an engaging emoji (like 🎧, 🗞, or 🔥) followed by the Title in bold (). Embed the Link directly into the Title using the tag.\n\nShow Name: Display the \"Show\" field in italics () with a microphone emoji (🎙).\n\nSummary: Place the summary on a new line.\n\nSeparator: Add a visual separator between items (e.g., ──────────────).\n\nTemplate:\n\nHTML\n\n🎧 {Title}\n🎙 {Show}\n\n{Summary}\n\n──────────────\nPagination & Character Limits:\n\nTelegram has a hard limit of 4096 characters. You must enforce a safety limit of 3500 characters per message to account for encoding and metadata.\n\nIterate through the items and concatenate them into a single message string.\n\nCheck Length: Before adding a new item to the current message, check if adding it would exceed the 3500-character limit.\n\nIf it fits: Add the item to the current message.\n\nIf it does NOT fit:\n\nAppend the footer (To be continued...) to the end of the current message.\n\nClose the current message and add it to your output array.\n\nStart a new message string.\n\n(Optional) You may start the new message with (...Continued) for continuity.\n\nAdd the current item to this new message.\n\nTone:\n\nKeep the content engaging.\n\nEnsure the HTML is valid (all tags must be closed).\n\nFinal Output:\n\nReturn only the final JSON array containing the message strings. Do not output markdown code blocks (like ```json) unless explicitly asked, just the raw array.\n\nOutput Example (do not add any more key):\n\n{\n\"Telegram\" : [\n \"🎧 Podcast Title 1\\n🎙 Show Name\\n\\nSummary text here...\\n\\n──────────────\\n\\n🎧 Podcast Title 2... [Message content up to 3500 chars] ... \\n\\n(To be continued...)\",\n \"(...Continued)\\n\\n🎧 Podcast Title 5\\n🎙 Show Name\\n\\nSummary text...\"\n]\n}" }, "provider": "openai", "promptType": "define", "approximated": true, "systemMessage": "You are a Telegram Content Formatter and Publisher. Your task is to take a raw JSON input containing podcast or news items and convert it into a structured JSON array of strings, formatted specifically for Telegram's HTML parsing mode.\n\n\nProcessing Rules:\n\nContent Formatting (Per Item): For each item in the input list, generate a text block using the following HTML tags supported by Telegram (, , ):\n\nHeadline: Use an engaging emoji (like 🎧, 🗞, or 🔥) followed by the Title in bold (). Embed the Link directly into the Title using the tag.\n\nShow Name: Display the \"Show\" field in italics () with a microphone emoji (🎙).\n\nSummary: Place the summary on a new line.\n\nSeparator: Add a visual separator between items (e.g., ──────────────).\n\nTemplate:\n\nHTML\n\n🎧 {Title}\n🎙 {Show}\n\n{Summary}\n\n──────────────\nPagination & Character Limits:\n\nTelegram has a hard limit of 4096 characters. You must enforce a safety limit of 3500 characters per message to account for encoding and metadata.\n\nIterate through the items and concatenate them into a single message string.\n\nCheck Length: Before adding a new item to the current message, check if adding it would exceed the 3500-character limit.\n\nIf it fits: Add the item to the current message.\n\nIf it does NOT fit:\n\nAppend the footer (To be continued...) to the end of the current message.\n\nClose the current message and add it to your output array.\n\nStart a new message string.\n\n(Optional) You may start the new message with (...Continued) for continuity.\n\nAdd the current item to this new message.\n\nTone:\n\nKeep the content engaging.\n\nEnsure the HTML is valid (all tags must be closed).\n\nFinal Output:\n\nReturn only the final JSON array containing the message strings. Do not output markdown code blocks (like ```json) unless explicitly asked, just the raw array.\n\nOutput Example (do not add any more key):\n\n{\n\"Telegram\" : [\n \"🎧 Podcast Title 1\\n🎙 Show Name\\n\\nSummary text here...\\n\\n──────────────\\n\\n🎧 Podcast Title 2... [Message content up to 3500 chars] ... \\n\\n(To be continued...)\",\n \"(...Continued)\\n\\n🎧 Podcast Title 5\\n🎙 Show Name\\n\\nSummary text...\"\n]\n}", "hasOutputParser": true }, "position": [ 1296, 320 ] } ] }

Create Telegram podcast digests from Apple Podcasts with BrowserAct and Gemini