Overview This n8n workflow automates the creation of 9:16 aspect ratio images optimized for short-form video content and thumbnails. It integrates multiple tools to retrieve content, generate scripts, and create AI-generated imagery. Key Features Trigger Workflow Manually The workflow starts when triggered manually in n8n. Retrieve Brand Guidelines Fetch brand elements like style, tone, and guidelines from Airtable. SEO Keywords and Blog Post Retrieval Retrieves blog posts and associated SEO keywords from Airtable to form the basis of image content. Content Preparation Uses GPT-4 to prepare a 4-scene script and thumbnail prompts for short-form videos. AI Image Generation Uses Leonardo.ai API to generate: Thumbnail Images Scene-specific Images (9:16 Aspect Ratio) Airtable Asset Management Generated assets (images) are saved back into Airtable with metadata like URLs and file sizes. Tools and Integrations n8n**: Workflow automation platform. OpenAI**: Generates scripts and prompts (GPT-4O-MINI). Leonardo.ai**: AI tool for improving prompts and generating high-quality images. Airtable**: Used as a data source for brand guidelines, blog posts, and to store generated assets.
Tags
AirtableWHTTP RequestOpenAI
WikipediaRelated workflows
See all AI Automation→BW
Send and track LinkedIn outreach with Airtable, Unipile, and OpenRouter
Quick overview This workflow automates LinkedIn outreach by using Airtable as a CRM, Unipile to fetch profiles and send connection requests/messages, and OpenRouter to generate personalized connection notes and follow-ups, while tracking accepts and replies via Unipile webhooks. How it works Runs every weekday at 09:00 to pull up to 15 Airtable leads with Connection Status set to New. For each lead, derives the LinkedIn public identifier, fetches the LinkedIn profile from Unipile, and uses OpenRouter to generate a personalized connection note capped at 300 characters. Sends the LinkedIn connection request via Unipile, updates the Airtable record with request details (note, timestamps, invitation ID), and waits 30–90 seconds between sends. Receives Unipile new_relation webhook events, matches the lead by Provider ID, and marks the Airtable record as Accepted with an Accepted At timestamp. Receives Unipile message_received webhook events, filters to inbound replies, finds the matching Airtable lead, and marks it Replied while stopping the campaign. Runs daily at 10:00 to find accepted, no-reply leads due for follow-up, uses OpenRouter to write follow-up #1 or #2, sends the DM via Uni
Create Zoho Books invoices from email attachments using OpenAI OCR
Quick overview This workflow monitors an IMAP inbox for emails with invoice attachments, uses OpenAI to OCR and extract structured invoice data, looks up customers and tax/currency settings in Zoho Books, and then creates a new Zoho Books invoice with the original attachment included. How it works Triggers when a new email arrives via IMAP. Filters for emails that contain attachments, then splits each attachment into its own item for processing. Normalizes each attachment’s binary data and routes it to OpenAI for OCR based on whether it is a PDF or an image. Fetches taxes, currencies, and active customer contacts from Zoho Books and prepares the data for matching and invoice creation. Sends the OCR text plus Zoho Books tax metadata to OpenAI to validate that the document is a sales invoice and to return invoice and customer details as JSON. Filters out non-invoice attachments, matches the extracted customer to an existing Zoho Books contact (with a configurable fallback), and posts a new invoice to Zoho Books using multipart upload to attach the original file. Setup Add IMAP credentials for the mailbox that receives invoice emails. Add Zoho Books OAuth2 credentials and ensure the c
Create Instagram carousels and TikTok slideshows with OpenAI and PostFast
Quick overview This workflow runs on a Mon/Wed/Fri schedule to generate an Instagram carousel or TikTok slideshow using OpenAI (GPT-4o-mini and gpt-image-1), sends the slides to Telegram for approval, and only then uploads the images and schedules the post via PostFast. How it works Runs every Monday, Wednesday, and Friday at 8am and loads your brand, topics, slide count, platform filter, and Telegram chat ID. Retrieves your connected social accounts from PostFast and keeps only accounts that are connected and match the selected platforms. Uses OpenAI (gpt-4o-mini) to write a carousel as strict JSON containing a caption and slide-by-slide headings, body text, and visual descriptions. Builds per-slide image prompts and uses OpenAI (gpt-image-1) to generate a 1024×1536 portrait image for each slide. Sends the generated slide images to Telegram and waits for you to approve or decline posting. If approved, requests per-account upload URLs from PostFast, uploads each slide image to the provided storage URLs, and schedules one PostFast social post per account with the slides in order and the generated caption. Setup Install and enable the PostFast community node (n8n-nodes-postfast) and