What's new
Bulk processing
Process multiple videos in one command:
/process https://youtube.com/watch?v=abc https://youtube.com/watch?v=def
Playlist support
Pass a playlist URL and all videos are processed automatically:
/process https://www.youtube.com/playlist?list=PLxxx
Already-processed videos are skipped.
Per-video file naming (breaking change for existing vaults)
All output files now include the video ID to prevent overwrites when multiple videos from the same channel are processed:
| Before | After |
|---|---|
transcript_raw.txt |
transcript_<video_id>.txt |
metadata.json |
metadata_<video_id>.json |
summary.md |
summary_<video_id>.md |
triplets.json |
triplets_<video_id>.json |
graph.json and graph.html remain shared per channel (cumulative graph).
How to update
git pull origin mainMigrate existing vault (optional)
If you have existing processed videos, rename your metadata.json files:
# example for one channel
mv vault/content/my_channel/raw/metadata.json vault/content/my_channel/raw/metadata_VIDEO_ID.jsonOr re-run python src/generate_video_db.py after renaming — it now scans metadata_*.json.