Skip to main content

youtube-to-writer

Scrapes a YouTube channel — current project is one channel, 876 videos and counting. Converts transcripts into a database, refines them through a learned keyword list and correction process, then researches across the full library to write new content — books, articles — in the speaker’s voice using a profile built from their existing published work. No database server, no orchestration. Four skills and a folder of markdown files.

SCRAPE            YouTube API → transcript extraction
                   (876 videos, growing)
──────────────────────────────────────────────────────
CLEAN             learned keyword list, correction rules
                   normalize names, terminology, references
──────────────────────────────────────────────────────
INDEX             categorize by topic, build keyword index
                   ← searchable across full library
──────────────────────────────────────────────────────
RESEARCH          cross-reference topics across 876 talks
                   assemble source material for new content
                   ← this is where the value is
──────────────────────────────────────────────────────
WRITE             voice profile from published work
                   → books, articles in the speaker's voice

Claude · markdown only