This release adds 2 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Topics
+8 more
Summary
AI summaryOptional YouTube vision indexing with on-screen enrichment via PINRAG_YT_VISION_* flags.
Full changelog
v0.9.24 — Optional YouTube vision indexing
Ships optional on-screen enrichment for YouTube (keyframes + vision model merged with transcript), pinrag[vision] and ffmpeg support, and updated docs and MCP marketplace metadata.
Features
- YouTube vision — Opt-in via
PINRAG_YT_VISION_*; scene detection, OpenAI or Anthropic vision, merged chunks with transcript; metadata includeshas_visual,frame_count,visual_source.
Docs & MCP
- README / LAUNCHGUIDE / env example — Install (
pinrag[vision]), ffmpeg, env table, DockerBUILD_WITH_VISION, re-index note. - server.json — Optional env vars for Anthropic and
PINRAG_YT_VISION_ENABLED; description shortened to satisfy MCP Registry length (≤100 characters).
Demo
- README — Demo video and assets refreshed in recent commits.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About ndjordjevic/pinrag
RAG for PDFs, YouTube, GitHub repos, Discord exports; index documents and query with citations.
Related context
Beta — feedback welcome: [email protected]