“Let's freeze the schema Friday - anything after that goes in the next migration window.”
Captured live by the assistant. Consent notice posted at join. Cited to the exact moment.
Paste a meeting link and the Citesvue assistant joins with a visible consent notice and records. The recap, transcript, transcript-derived findings, and transcript Q&A arrive first; optional visual evidence can be requested later. Upload and bot capture follow the same flow.
“Let's freeze the schema Friday - anything after that goes in the next migration window.”
Captured live by the assistant. Consent notice posted at join. Cited to the exact moment.
Drop a meeting link into Citesvue - Meet, Teams, Zoom, Webex, or Lark - from your calendar view or straight from the clipboard.
The assistant joins as a visible participant with a clear bot name and posts a consent notice where the platform allows it.
It records the available meeting media. Core transcript processing starts after capture; for eligible video, a workspace member can later request visual evidence when screen proof is needed.
Minutes after the call: recap, transcript with speakers, transcript findings, and Q&A; visual evidence remains optional.
Paste a link from any of these and the assistant takes it from there. Anything not on this list yet is simply not on it - no asterisks.
| Platform | Status | Good to know |
|---|---|---|
| Available | Nothing to set up - paste the link and it joins. | |
| Available | Organisation-locked meetings may need the organiser to admit it from the lobby. | |
| Available | The meeting host approves recording when the assistant asks. | |
| Available | Transcripts are captured in full; per-speaker labels are coming to Webex. | |
| Available | Nothing to set up - paste the link and it joins. | |
| Available | Recorded natively by Nexynt - nothing extra joins the call. |
Any platform, any time: recordings from tools we don't join - Zoom local recordings, Loom, OBS, QuickTime - can always be uploaded directly and are processed identically.
The assistant is an ingestion path, not a separate product. Core transcript features arrive first; video-specific evidence remains optional.
Word-level timestamps, speaker labels you can correct once and keep.
Transcript-derived outcomes and artifacts are available without visual analysis.
Request scenes, OCR, UI state, and error-screen analysis when the call contains useful video.
Ask from the transcript immediately; visual citations join future answers only after evidence is ready.
Google Calendar connect is in early access, with Microsoft 365 next - connected calendars show your upcoming meetings with a Join Now button on each. Calendar-driven auto-join - the assistant sending itself to the meetings you choose - is in closed beta and opens up gradually. This page will say so plainly when it does.
Bot minutes come from the same plan catalogue the product enforces - the number below is the number you get.
Concurrent live-capture capacity is set by plan: the pricing table shows the enforced workspace limit, and requests beyond that capacity wait for an available slot. Free plan does not include the assistant - uploads work on every plan. Full pricing →
The findings pipeline every capture flows into.
Question the call afterwards; every answer cites its moment.
The step-by-step: paste the link, admit the assistant, review the record.
Where a citation-first assistant differs from a summary-first one.
Or it could sit in a Drive folder nobody opens again. The difference is whether it has citations attached.