jimmylv/bibi
Summarize, transcribe and search YouTube, Bilibili, TikTok, podcasts and local videos from your agent: timestamped transcripts, chapter summaries, saved-video library, channel feeds and on-screen vision via the hosted BibiGPT MCP server and 4 skills.
Video summarizer agent skill for Claude Code, Codex, ChatGPT, Cursor and OpenClaw. Summarize YouTube, Bilibili, TikTok, Twitter/X, Xiaohongshu (video and image notes), podcasts and any online or local video / audio; extract timestamped transcripts and subtitles; chapter-by-chapter summaries; optionally remix the takeaway into a TikTok-style vertical music video. Not for already-saved videos (bibi-library), channel feeds (bibi-feed), or slides / frames / mind maps (bibi-vision): sibling skills, same CLI and account. Triggers: "summarize this video", "what's this video about", "get transcript", "extract subtitles", "YouTube summary", "podcast notes", "video to notes", "video summarizer", "总结这个视频", "帮我看看这个视频讲了什么", "B站总结", "video to TikTok MV", "把视频变成 TikTok". Works via the bibi CLI (macOS/Windows), the hosted MCP server, or OpenAPI (Linux / no CLI).
Subscribe to YouTube / Bilibili / podcast channels in BibiGPT, pull the latest feed, and mark items seen. Use when the user wants channel subscriptions, "what's new in my feed", a daily digest, or to subscribe / unsubscribe. Do not use to summarize a new URL (bibi), search already saved videos (bibi-library), or analyze frames / mind maps (bibi-vision). Triggers: "subscribe to this channel", "what's new in my subscriptions", "daily digest", "mark feed seen", "订阅频道", "我的订阅", "有什么更新", "取消订阅", "latest feed". Same bibi CLI binary and same BibiGPT account as the other skills.
Browse, search, and manage a BibiGPT saved library: already-summarized videos, personal notes, and collections. Use when the user asks what they already saved, to search past summaries, edit a note, or organize videos into a collection. Do not use to summarize a new URL (that is the bibi skill), follow channels / pull a subscription feed (bibi-feed), or analyze frames / generate a mind map (bibi-vision). Triggers: "what have I summarized", "search my library", "my saved videos", "my notes", "create a collection", "add to collection", "我之前总结过的", "资料库", "搜我的笔记", "合集", "列出最近的总结". Same bibi CLI binary and same BibiGPT account as the other skills.
Extract visual content from a BibiGPT video (slides, OCR, on-screen text) and generate a mind map from a saved summary. Use when the user asks what is on screen, to analyze slides / PPT / frames, or to make an XMind mind map. Do not use to summarize audio or get a transcript (bibi), search the saved library (bibi-library), or manage channel feeds (bibi-feed). Do not invent digital humans, lip-sync, or shaders — BibiGPT does not expose those. Triggers: "what's on screen", "analyze the slides", "visual analysis", "画面分析", "PPT 上写了什么", "mind map", "思维导图", "extract visuals". Same bibi CLI binary and same BibiGPT account as the other skills.