List every video on a channel or a playlist: each upload of a YouTube channel (long videos, shorts and live streams together), a channel's tab, or any playlist, with its link, title, length, views, kind and thumbnail, read on the person's own computer
Runs on the person's computer: lists the videos at url (a channel, a channel's tab or a playlist) with the computer's link reader (the hear-fetcher rows with a list command, such as yt-dlp), without reading each video. The video-list-rule rows add what a page needs beyond one read: for a YouTube channel, its uploads list (every long video and short together, with the count the channel states) and its own page (which tells which videos are shorts). Answers { url, title, channel, channel_id, count, stated_count, complete, entries: [{ id, url, title, duration_seconds, view_count, kind: video, short, live or null, thumbnail }], unread }. complete says whether count reaches the count the page states; unread names any added list that could not be read. A link that lists nothing fails with the reason, never an empty answer.
Connect the AI you already use to Trillion once. Then ask in your own words; your AI finds this tool by what it does and runs it.
Say to your AI: “Use Trillion to list every video on a channel or a playlist: each upload of a YouTube channel (long videos, shorts and live streams together), a channel's tab, or any playlist, with its link, title, length, views, kind and thumbnail, read on the person's own computer”
What your AI calls: run_tool {"tool_id":"tool:list-every-video-on-a-channel"}