Daily News In, Ready-to-Publish Posts Out

The problem

A parenting content brand needed a daily stream of research-backed Facebook posts for a Traditional Chinese-speaking audience. Someone had to read the day’s parenting news across multiple sources, summarise it, and rewrite it into the brand’s structured post format — every single day.

The build

I built an n8n workflow that runs on a schedule and does the entire pipeline end to end. It reads the list of active sources from a Google Sheet — RSS feeds and plain web pages alike — fetches each one, and keeps only articles published in the last 24 hours.

RSS items go through a feed parser; plain pages get fetched, stripped of HTML, and reduced to clean readable text, trimmed to keep AI token costs predictable. Each article then goes to GPT-4o-mini with a tightly constrained prompt that returns three outputs in Traditional Chinese: a 300–400 word Facebook post following the brand’s structure (hook, findings, parent perspective, actionable tips), a one-sentence CTA, and an image-generation prompt for branded graphics.

Results land in an output Google Sheet where the content team reviews and schedules them.

The result

A fully automated daily content pipeline — from raw news to publish-ready posts — with the team only doing final review. The client came back with a second automation order the same week.

“Exceptional services and incredible efficiency.” — repeat client, via Fiverr