tunnel_url string | mcp_url string | sse_url string | updated_at float64 |
|---|---|---|---|
https://procurement-rated-fort-called.trycloudflare.com | https://procurement-rated-fort-called.trycloudflare.com/mcp | https://procurement-rated-fort-called.trycloudflare.com/sse | 1,793,081,088.269466 |
Sadhguru Archive — Transcript & Vector Index Dataset
This dataset powers the Sadhguru Archive Studio & MCP Server (videopublication/sadhguru-archive).
Included Files
sadhguru.db: SQLite database (~455 MB) with FTS5 full-text index, video metadata, chapters, word-timed ASR transcript lines (lines), and official YouTube closed captions (captions) across ~5,200+ talks, shorts, and live streams.embeddings_m3.npz:BAAI/bge-m3(1024-dim float16) semantic passage embeddings for ASR transcripts.cc_embeddings_m3.npz:BAAI/bge-m3(1024-dim float16) semantic passage embeddings for YouTube CC tracks.
Automatic Local Sync
When you clone the repository and run the app, these files are automatically downloaded into ./data (or SADHGURU_ARCHIVE_DIR):
python -m archive.cloud_sync --pull
- Downloads last month
- 1,093