CORTEXFLOW · FREE FOREVER

Automation

AI Devotional Reels Factory

↓ Download workflow JSON

Import in n8n: Workflows → ⋯ → Import from file, then add your credentials. Free download with a free account.

How it works

Send a deity or scene photo with a caption idea to your Telegram bot — the workflow does the rest:

  1. Telegram Trigger collects the idea — every photo message (with caption) fires the workflow via webhook.
  2. The photo is archived — the file is pulled from Telegram, uploaded to Google Drive, and logged in Google Sheets with status RECEIVED.
  3. GPT-4o Vision analyzes the photo — it outputs a structured YAML description (subject, colors, setting, mood) so every later step stays faithful to the reference.
  4. NanoBanana creates the keyframe — an AI agent writes a devotional keyframe prompt (diya glow, marigold tones, cinematic realism), fal.ai renders it, and the result is downloaded.
  5. An AI agent writes the reel script — using a structured master prompt, it outputs a title plus a full VEO3-ready JSON prompt (camera, lighting, motion, VFX, audio).
  6. VEO3 generates the vertical video — kie.ai renders a 9:16 reel from the keyframe image, then the workflow polls until it’s ready.
  7. GPT-4o writes the Hindi caption — a reverent Devanagari caption under 150 characters with bhakti hashtags.
  8. Blotato posts everywhere — the video is uploaded once and published to Instagram, YouTube Shorts, TikTok, Facebook, and X in parallel; the sheet flips to Published and you get a Telegram confirmation.

Set up steps

You’ll need:

  • n8n (self-hosted or cloud) with webhook access
  • A Telegram bot token — message @BotFather, send /newbot, copy the token
  • An OpenAI API key — platform.openai.com → API keys
  • A kie.ai API key — kie.ai → API keys (for VEO3)
  • A fal.ai API key — fal.ai → Keys (for NanoBanana)
  • A Blotato account with API access — blotato.com → Settings → API, and connect your Instagram, YouTube, TikTok, Facebook, and X accounts there
  • A Google account — for Drive and Sheets (OAuth sign-in inside n8n)
  1. Download the workflow JSON using the download button on this page.
  2. In n8n, go to Workflows → ⋯ → Import from file and select the JSON.
  3. Telegram — Credentials → New → Telegram, paste your bot token. Select it in Telegram Trigger: Receive Reel Idea, Telegram: Get Photo File, Send Final Reel Preview, and Telegram: Send Published Notification. Then open the Set: Bot Token node and paste the same token into YOUR_BOT_TOKEN.
  4. OpenAI — Credentials → New → OpenAI, paste your API key. Select it in OpenAI Vision: Analyze Photo, LLM: OpenAI Chat, OpenAI Chat Model, and Rewrite Caption with GPT-4o.
  5. kie.ai — Credentials → New → Header Auth, name it kie.ai API Key, set Name to Authorization and Value to Bearer YOUR_KIE_KEY. Select it in Generate Video with VEO3 and Download Video from VEO3.
  6. fal.ai — Credentials → New → Header Auth, name it fal.ai API Key, set Name to Authorization and Value to Key YOUR_FAL_KEY. Select it in NanoBanana: Create Image and Download Edited Image.
  7. Google — create a Google Drive OAuth2 credential and a Google Sheets OAuth2 credential (n8n walks you through Google sign-in). Select them in the Drive/Sheets nodes. Create a tracker sheet with columns IMAGE NAME, IMAGE URL, IMAGE DESCRIPTION, CAPTION, TITRE VIDEO, CAPTION VIDEO, URL VIDEO FINAL, STATUS, then open each Google Sheets node and pick your document and sheet.
  8. Blotato — in n8n, enable Verified Community Nodes under Settings, install the Blotato node, then Credentials → New → Blotato and paste your API key. Select it in Upload Video to BLOTATO and each platform node, then pick the right social account in every platform node.
  9. Activate the workflow, then send your bot a photo with a caption like “Radha Krishna at Vrindavan ghat at dawn”.

Customize it

  • More platforms — duplicate any Blotato platform node, pick another platform (LinkedIn, Threads, Bluesky, Pinterest), and wire it into the Merge node (raise Number of inputs to match).
  • Different video ratio — edit the Format Prompt code node: change aspectRatio to "16:9" for YouTube-style landscape.
  • Caption language — edit the prompt in Rewrite Caption with GPT-4o to write in English or Hinglish instead of Hindi.
  • Faster/cheaper video — in Format Prompt, change model to another kie.ai video model.
  • Approval before posting — insert a Wait (webhook) or Telegram approval step between Send Final Reel Preview and Upload Video to BLOTATO so nothing posts without your OK.
  • Posting schedule — add a Schedule Trigger branch that reads READY rows from the sheet and posts them at fixed times instead of instantly.

Workflow nodes

This workflow is built from 36 nodes. Here's what each one does:

#NodeTypeWhat it does
1 Telegram Trigger: Receive Reel Idea Telegram Trigger Starts the workflow when a Telegram message arrives.
2 Telegram: Get Photo File Telegram Sends you a Telegram message or alert.
3 Google Drive: Upload Photo Google Drive Uploads, downloads, or manages files in Google Drive.
4 Google Sheets: Log Idea Google Sheets Reads from or writes to a Google Sheet — this workflow's log and state store.
5 Set: Bot Token Set Holds editable configuration values — fill in your own settings here before activating.
6 Telegram API: Get File URL HTTP Request Calls an external API or webhook endpoint.
7 OpenAI Vision: Analyze Photo OpenAI (sub-node) OpenAI sub-node attached to an AI agent.
8 Google Sheets: Save Photo Description Google Sheets Reads from or writes to a Google Sheet — this workflow's log and state store.
9 LLM: OpenAI Chat OpenAI Chat Model The language model powering the AI agent.
10 LLM: Structured Output Parser Structured Output Parser Forces the AI to return clean, structured JSON instead of free text.
11 Generate Image Prompt AI Agent The AI brain — reasons over the task and calls tools to get it done.
12 NanoBanana: Create Image HTTP Request Calls an external API or webhook endpoint.
13 Wait for Image Edit Wait Pauses the workflow until a condition or time is met.
14 Download Edited Image HTTP Request Calls an external API or webhook endpoint.
15 Google Sheets: Read Reel Config Google Sheets Reads from or writes to a Google Sheet — this workflow's log and state store.
16 Set: Master Prompt Set Holds editable configuration values — fill in your own settings here before activating.
17 OpenAI Chat Model OpenAI Chat Model The language model powering the AI agent.
18 Think Think Tool Lets the agent 'think' step-by-step before acting.
19 Structured Output Parser Structured Output Parser Forces the AI to return clean, structured JSON instead of free text.
20 AI Agent: Generate Reel Script AI Agent The AI brain — reasons over the task and calls tools to get it done.
21 Format Prompt Code Runs custom JavaScript to transform, filter, or compute data between steps.
22 Generate Video with VEO3 HTTP Request Calls an external API or webhook endpoint.
23 Wait for VEO3 Rendering Wait Pauses the workflow until a condition or time is met.
24 Download Video from VEO3 HTTP Request Calls an external API or webhook endpoint.
25 Rewrite Caption with GPT-4o OpenAI (sub-node) OpenAI sub-node attached to an AI agent.
26 Save Reel to Google Sheets Google Sheets Reads from or writes to a Google Sheet — this workflow's log and state store.
27 Send Final Reel Preview Telegram Sends you a Telegram message or alert.
28 Upload Video to BLOTATO Blotato (Social Poster) Publishes posts to social platforms via Blotato.
29 Instagram Blotato (Social Poster) Publishes posts to social platforms via Blotato.
30 YouTube Blotato (Social Poster) Publishes posts to social platforms via Blotato.
31 TikTok Blotato (Social Poster) Publishes posts to social platforms via Blotato.
32 Facebook Blotato (Social Poster) Publishes posts to social platforms via Blotato.
33 Twitter (X) Blotato (Social Poster) Publishes posts to social platforms via Blotato.
34 Merge Merge Combines the outputs of two or more branches back into one flow.
35 Update Status to Published Google Sheets Reads from or writes to a Google Sheet — this workflow's log and state store.
36 Telegram: Send Published Notification Telegram Sends you a Telegram message or alert.

Setup instructions

1. Add your credentials

In n8n go to Credentials → New, create each credential below, then open the named node and select it:

NodeCredential to add
Telegram Trigger: Receive Reel Idea Telegram Bot API
Telegram: Get Photo File Telegram Bot API
Google Drive: Upload Photo Google Drive OAuth2
Google Sheets: Log Idea Google Sheets OAuth2
OpenAI Vision: Analyze Photo OpenAI API
Google Sheets: Save Photo Description Google Sheets OAuth2
LLM: OpenAI Chat OpenAI API
NanoBanana: Create Image HTTP Header Auth
Download Edited Image HTTP Header Auth
Google Sheets: Read Reel Config Google Sheets OAuth2
OpenAI Chat Model OpenAI API
Generate Video with VEO3 HTTP Header Auth
Download Video from VEO3 HTTP Header Auth
Rewrite Caption with GPT-4o OpenAI API
Save Reel to Google Sheets Google Sheets OAuth2
Send Final Reel Preview Telegram Bot API
Upload Video to BLOTATO Blotato API
Instagram Blotato API
YouTube Blotato API
TikTok Blotato API
Facebook Blotato API
Twitter (X) Blotato API
Update Status to Published Google Sheets OAuth2
Telegram: Send Published Notification Telegram Bot API

How this workflow starts

  • Telegram Trigger: Fires automatically on its event — activate the workflow and connect the service.

Use cases

Send a deity or scene photo with a caption idea to a Telegram bot, and this workflow builds a cinematic devotional reel end to end. The photo is pulled from Telegram, archived to Google Drive, and logged in Sheets with status RECEIVED. OpenAI Vision: Analyze Photo produces a structured YAML description so every later step stays faithful to the reference. An agent writes the NanoBanana keyframe prompt, fal.ai renders the keyframe, AI Agent: Generate Reel Script outputs a VEO3-ready JSON prompt covering camera, lighting, motion, VFX, and audio, and kie.ai's VEO3 renders the 9:16 vertical video with polling waits until it is ready. Rewrite Caption with GPT-4o writes a reverent Devanagari caption under 150 characters with bhakti hashtags, and Blotato publishes to Instagram, YouTube Shorts, TikTok, Facebook, and X in parallel before the sheet flips to Published and a Telegram confirmation arrives. It is built for devotional creators and temple teams who want daily cinematic reels without an editing desk.

The temple trust posting daily darshan reels

A temple trust's media volunteer is expected to post a reel every morning, but he has no editing skills and a full-time job. His new workflow is taking one photo at the temple and sending it to the bot with a caption idea like "morning aarti diyas." The pipeline handles the rest: keyframe, script, VEO3 render, Hindi caption, five-platform publish. The Google Sheets: Log Idea tracker, with its STATUS column moving from RECEIVED to Published, gives the trust committee a visible content calendar without anyone asking for reports. Practical variation: add the posting-schedule customization — a Schedule Trigger branch that reads READY rows from the sheet and posts at fixed times — so festival-day reels queue up instead of publishing the instant they render.

The solo devotional creator running five platforms from a phone

A creator with eighty thousand followers was reformatting and uploading the same video five times, losing an hour a day to busywork. The Upload Video to BLOTATO step plus the five Blotato platform nodes (Instagram, YouTube, TikTok, Facebook, Twitter/X) merged back together publish everywhere from a single upload, and Telegram: Send Published Notification confirms it landed. Her entire production loop is now: photograph, caption idea, wait, confirm. Practical tip: insert the approval customization — a Wait or Telegram approval step between Send Final Reel Preview and Upload Video to BLOTATO — so nothing goes public before she has seen the final render on her phone.

The spiritual publisher mining a photo archive

A publisher sits on ten thousand festival photographs that have never been used for video. The archive path is the Google Drive: Upload Photo node paired with the tracker sheet's IMAGE URL and IMAGE DESCRIPTION columns: every photo becomes a catalogued asset, and OpenAI Vision: Analyze Photo writes the structured description (subject, colors, setting, mood) that keeps the NanoBanana keyframe faithful to the original scene rather than drifting into generic imagery. Each archived photo can then be re-run through the reel pipeline on festival anniversaries. Practical variation: edit the Format Prompt code node to change aspectRatio to 16:9 for YouTube-style landscape versions of the same reels.

The bhakti community manager with zero video skills

A community manager for a 40,000-member devotional group was asked to "do reels" with no budget and no editor. Her technical ceiling is sending a Telegram message, which is exactly the workflow's input: photo plus a caption idea in plain words. The Generate Image Prompt agent translates her simple idea into the devotional keyframe language (diya glow, marigold tones, cinematic realism) she could never write herself. Practical tip: if her audience reads Hinglish better than pure Hindi, edit the prompt in Rewrite Caption with GPT-4o to write Hinglish captions instead — one prompt change, every future reel follows.

The overseas devotee page keeping festivals alive

A diaspora-run page wants to post through every major festival but its admins are spread across three time zones and keep missing days. The workflow's structure solves the coordination problem: anyone can send a photo idea to the bot at any hour, the Wait for VEO3 Rendering node handles the slow render step without babysitting, and the sheet's URL VIDEO FINAL column keeps every finished reel in one place for resharing next year. Practical variation: duplicate any Blotato platform node, pick Pinterest or Threads, and wire it into the Merge node (raising Number of inputs to match) to reach the diaspora audiences that live on those platforms.

← All templates