Search, download, upload, and edit Feishu Minutes recordings via lark-cli.
npx skills add https://github.com/TrayMachi/dotfiles --skill lark-minutes-traymachi
Please help me install this Agent Skill. Skill: lark-minutes Source: https://github.com/TrayMachi/dotfiles/tree/main/agents/skills/lark-minutes Command: npx skills add https://github.com/TrayMachi/dotfiles --skill lark-minutes-traymachi
minutes get. - Media Download & Upload: Download minute audio/video files (or get 1-day shareable links), and convert local media files into minutes through the drive +upload → minutes +upload → vc +notes pipeline instead of local transcription tools like ffmpeg or whisper. - Edit AI Artifacts: Rename minutes, replace transcript speakers, batch add/update/delete AI todos inside a minute, and replace the full AI summary. - Use Case: A user says "turn this mp3 recording into a transcript with a summary" — the skill uploads the file to Feishu Drive, generates a minute, then fetches the transcript and AI summary via vc +notes. ## Quick Start Ask the assistant to search your Feishu minutes from last week and download the matching recording to a local folder.High-intent search queries and answers about installing and using this skill.
Use `lark-cli minutes +search` with `--query`, `--start`/`--end`, `--owner-ids`, or `--participant-ids` filters. At least one filter is required, and results paginate via `page_token` when `has_more` is true.
Upload the file with `lark-cli drive +upload` to get a file_token, run `lark-cli minutes +upload --file-token` to generate the minute, then extract the minute_token from the returned URL and call `lark-cli vc +notes --minute-tokens` for the transcript and summary.
Yes, `lark-cli minutes +download --minute-tokens <token>` saves the media locally, or use `--url-only` for a download link valid for one day. The minute must have finished transcription, and the minutes:minutes.media:export scope is required.
Supported audio formats include wav, mp3, m4a, aac, ogg, wma, and amr; video formats include avi, wmv, mov, mp4, m4v, mpeg, ogg, and flv. Files must be under 6 hours long and 6 GB in size.
The `+speaker-replace` command only accepts `ou_` prefixed open_ids, not display names. Resolve names to open_ids via the contact skill first, and ensure the from-user already exists as a speaker in that minute's transcript.
Do not use it to read transcripts, summaries, or chapters of an existing minute — that belongs to `vc +notes` in the lark-vc skill. Also avoid it for locating notes by natural-language title only, which routes to drive or doc search instead.
DeepJHSearch, download, and edit Feishu Minutes recordings and AI-generated meeting artifacts.
trungpr1990-coderSearch Lark video meeting history and retrieve meeting notes, transcripts, and participant snapshots.
agent-pilotFetch historical video conference records and retrieve notes and transcripts.
KaiBoo404Fetch YouTube transcripts and captions using yt-dlp, optionally via Whisper, producing .vtt files.