mcptube

is this based on Karapathy's LLM Wiki idea (link: https://gist.github.com/karpathy/442a6bf555914893e9891c11519...)? . I leveraged Karapathy's wiki idea and built MCPTube- it's a CLI and also an MCP server that turns YouTube videos into a compounding knowledge base. Check it out and let me know what you think (link: https://github.com/0xchamin/mcptube)

I watch a lot of Stanford/Berkeley lectures and YouTube content on AI agents, MCP, and security. Got tired of scrubbing through hour-long videos to find one explanation. Built v1 of mcptube a few months ago. It performs transcript search and implements Q&A as an MCP server. It got traction (34 stars, my first open-source PR, some notable stargazers like CEO of Trail of Bits). But v1 re-searched raw chunks from scratch every query. So I rebuilt it. v2 (mcptube-vision) follows Karpathy's LLM Wiki pattern. At ingest time, it extracts transcripts, detects scene changes with ffmpeg, describes key frames via a vision model, and writes structured wiki pages. Knowledge compounds across videos rather than being re-discovered. FTS5 + a two-stage agent (narrow then reason) for retrieval. MCPTube works both as CLI (BYOK) and MCP server. I tested MCPT

I built mcptube because I spend a lot of time learning from Stanford and Berkeley lectures on YouTube, and I wanted a better way to interact with the content. mcptube ingests YouTube videos (transcripts, metadata, frames) and exposes them through both a CLI and an MCP server. You can semantically search transcripts, extract frames at any moment, ask questions about videos, and generate illustrated reports. It works with Claude Code, VS Code Copilot, Claude Desktop, and any MCP-compatible client. The MCP tools are passthrough — the connected LLM does the heavy lifting, zero API key needed on the server. The CLI also supports BYOK for more deterministic results. Tech stack: FastMCP, yt-dlp, ChromaDB, LiteLLM, SQLite. GitHub: https://github.com/0xchamin/mcptube PyPI: https://pypi.org/project/mcptube/ pipx install mcptube --python python3.12