n8n RSS to LinkedIn Post Automation: OpenAI, Gmail & Sheets

Automatically turn RSS news into a polished LinkedIn post—then get your approval before it publishes

This n8n workflow reads daily RSS headlines, uses OpenAI to draft a new LinkedIn post in Brazilian Portuguese, finds a matching image via Pexels, and emails you a ready-to-post version for approval. Only after you approve will it publish to your LinkedIn profile and log the result in Google Sheets.

What this workflow does

• Runs daily at 08:00 (schedule is configurable in n8n).
• Reads your recent post history from Google Sheets to provide context and avoid repeating topics.
• Fetches the latest RSS items from a predefined list of RSS feeds.
• Generates a LinkedIn post with OpenAI: sends RSS headlines/summaries to OpenAI and receives a structured JSON output (e.g., topic, post text, theme, and a source link).
• Selects a relevant image from Pexels by querying the Pexels API using the generated theme/topic and retrieving an image URL.
• Emails you a draft for approval via Gmail, including the post and image link.
• Supports an approval/rewrite loop:
  • If you request a rewrite, your feedback is sent back to OpenAI to surgically edit the post.
  • Then it re-parses the JSON output and asks for approval again.
• Publishes after approval: downloads the selected image, uploads it to LinkedIn, posts to your profile, and logs the outcome (or logs rejection) in Google Sheets.

Use cases

• Consistently publishing on LinkedIn using current industry news without manual drafting.
• SaaS operators who want a daily content pipeline with human approval to maintain tone and accuracy.
• Automation engineers building an RSS-to-social workflow with OpenAI + review gates.

Technical details

• n8n nodes/integrations called out in this workflow: Code, RSS Feed Read, OpenAI, HTTP Request, Pexels, Gmail, Switch, and supportive nodes like Sticky Note.
• Requires credentials for Google Sheets, OpenAI, and Gmail, plus the Google Sheets document ID and sheet name used for reading history and logging outcomes.