Rebuilding johnliter.blog: Turning My Blog Into an Automated Publishing Hub

A developer rebuilt their personal blog to eliminate duplicate publishing work by turning it into an automated hub that syncs Substack content via RSS. The new stack uses n8n to poll the Substack RSS feed on a schedule, extract article metadata (title, excerpt, image URL, GUID), and upsert records into MySQL. A PHP API serves that data to a React frontend. The old site stored full article bodies with voting, view tracking, and Markdown rendering — all removed in favor of a lean metadata-only schema. The source_guid field from the RSS feed prevents duplicate inserts. A second n8n step auto-marks the newest article as featured. The redesigned frontend uses a dark editorial aesthetic with image cards that include title, excerpt, date, and category. The architecture keeps n8n invisible to the browser — React only talks to a PHP API endpoint, while n8n runs in the background.