Plaud

Get current authenticated user info

get_current_user

Get the full record for a Plaud recording: metadata, the audio download URL, and an inventory of its transcript blocks and notes. This is the tool that returns audio — `presigned_url` is a signed link to the recording's audio file, valid 24h, and is how you download it or hand the user a download link. Call this whenever the user asks for the audio, the download link, or the recording file itself; use get_transcript or get_note to read transcript or note bodies instead.

get_file

Fetch the notes on a Plaud recording — one entry per tab in the app, so usually more than one: the AI summary (action items, key topics), any template tab, a saved Ask Plaud answer, and the highlights note when the user pressed the highlight button while recording.

get_note

Fetch the timestamped transcript with speaker attribution for a Plaud recording. Defaults to the `transaction` block (real speaker names + timestamps), returned one page of utterances at a time to stay within client size limits — call again with the returned `next_cursor` to fetch the next page. Set `block` to `outline`, `transaction_polish`, or `mark_memo` (the moments the user flagged with the device's highlight button) to fetch those blocks instead.

get_transcript

List Plaud recordings. Supports optional client-side filtering: `query` (case-insensitive name substring), `date_from`/`date_to` (YYYY-MM-DD, inclusive). Filtering scans the 500 most recent recordings and matches within them — it is a scan budget, not a match budget. Check `complete` before reporting a result as exhaustive: when it is false the `note` says how far back the scan reached, and an empty `data` may simply mean the requested range was never searched.

list_files