API

Every tldv.api.* operation is listed below with parameter shapes and return types from the plugin Zod schemas.

New to Corsair? See API access, authentication, and error handling.

Folders

list

folders.list Search the folders a user can see, with their meeting ids Risk: read

await corsair.tldv.api.folders.list({ query: 'Q3 renewals', limit: 20 });

Input

Name	Type	Required	Description
`query`	`string`	No	Folder name to search for
`meetingIds`	`string[]`	No	Folders must contain these meeting ids
`meetingIdsOp`	`and \| or`	No	Whether a folder must contain every id or any
`space`	`all \| personal \| team \| organization`	No	Which folder scopes to search
`page`	`number`	No	—
`limit`	`number`	No	—

Answers { page, pages, total, pageSize, results }, where each folder carries its meetingIds. space narrows the search to personal, team or organization folders. Output

Name	Type	Required	Description
`page`	`number`	Yes	Current page number
`pages`	`number`	Yes	Total pages available
`total`	`number`	Yes	Total results available
`pageSize`	`number`	Yes	Results per page
`results`	`object[]`	Yes	—

results full type

{
  id: string,
  name: string,
  meetingIds: string[]
}[]

Meetings

get

meetings.get Get one meeting by id, including the phone number and metadata an import attached Risk: read

await corsair.tldv.api.meetings.get({ meetingId: 'meeting-123' });

Input

Name	Type	Required	Description
`meetingId`	`string`	Yes	—

phoneNumber and metadata are populated here and are absent from the list endpoint; both are set when the meeting was imported or uploaded. Output

Name	Type	Required	Description
`id`	`string`	Yes	—
`name`	`string`	Yes	—
`happenedAt`	`string`	Yes	When the meeting happened
`url`	`string`	Yes	Console URL for the meeting
`duration`	`number`	Yes	Length in seconds
`organizer`	`object`	Yes	—
`invitees`	`object[]`	Yes	—
`template`	`object \| null`	No	—
`extraProperties`	`object \| null`	No	—
`phoneNumber`	`string \| null`	No	Only on a single-meeting read, not the list
`metadata`	`Record<string, any> \| null`	No	Only on a single-meeting read, not the list

organizer full type

{
  name?: string | null,
  email: string
}

invitees full type

{
  name?: string | null,
  email: string
}[]

template full type

{
  id?: string | null,
  label?: string | null
} | null

extraProperties full type

{
  conferenceId?: string | null
} | null

import

meetings.import Import a meeting from a public media URL, which starts a processing job and spends recording minutes Risk: write

await corsair.tldv.api.meetings.import({ url: 'https://example.org/call-recording.mp4', name: 'Discovery call', participants: ['jordan@example.org'] });

Input

Name	Type	Required	Description
`url`	`string`	Yes	Publicly reachable media URL. Supported: .mp3 .mp4 .wav .m4a .mkv .mov .avi .wma .flac
`name`	`string`	Yes	Name for the new meeting
`happenedAt`	`string`	No	Meeting date, ISO 8601 with milliseconds
`participants`	`string[]`	No	Emails of the invited participants
`phoneNumber`	`string`	No	—
`metadata`	`Record<string, string \| number \| boolean>`	No	Flat key/value map with your own identifiers: up to 20 keys, values are strings, numbers or booleans

Starts a processing job and answers { success, jobId, message }. url and name are both required. The URL must be publicly reachable and the media must be one of .mp3 .mp4 .wav .m4a .mkv .mov .avi .wma .flac. metadata is returned verbatim on the meeting and in the MeetingReady webhook, which is how you correlate your own record ids. It takes up to 20 keys, and each value is a string of up to 256 characters, a number or a boolean. Output

Name	Type	Required	Description
`success`	`boolean`	Yes	—
`jobId`	`string`	Yes	Id of the import job
`message`	`string`	Yes	—

list

meetings.list List meetings with paging, date bounds and folder or type filters Risk: read

await corsair.tldv.api.meetings.list({ limit: 10, meetingType: 'external' });

Input

Name	Type	Required	Description
`query`	`string`	No	Search text
`page`	`number`	No	Page number, from 1
`limit`	`number`	No	Results per page, max 100
`from`	`string`	No	Earliest date, as a date or date-time
`to`	`string`	No	Latest date, as a date or date-time
`onlyParticipated`	`boolean`	No	Only meetings the user took part in
`meetingType`	`internal \| external`	No	internal means only participants from the org
`folderIds`	`string[]`	No	Folders to include
`folderIdsOp`	`and \| or`	No	Whether a meeting must be in every folder or any

Answers { page, pages, total, pageSize, results }. limit caps at 100. meetingType: 'internal' keeps only meetings whose participants are all in your organization; 'external' keeps the rest. from and to accept a date or a date-time. The input schema rejects a meetingType other than these two before any request is sent. query matches whole words: a meeting named weekly-sync is found by weekly sync, but not by weekly-sync. Output

Name	Type	Required	Description
`page`	`number`	Yes	Current page number
`pages`	`number`	Yes	Total pages available
`total`	`number`	Yes	Total results available
`pageSize`	`number`	Yes	Results per page
`results`	`object[]`	Yes	—

results full type

{
  id: string,
  name: string,
  happenedAt: string,
  url: string,
  duration: number,
  organizer: {
    name?: string | null,
    email: string
  },
  invitees: {
    name?: string | null,
    email: string
  }[],
  template?: {
    id?: string | null,
    label?: string | null
  } | null,
  extraProperties?: {
    conferenceId?: string | null
  } | null,
  phoneNumber?: string | null,
  metadata?: Record<string, any> | null
}[]

notes

meetings.notes Get the AI notes and topics generated for one meeting Risk: read

await corsair.tldv.api.meetings.notes({ meetingId: 'meeting-123' });

Input

Name	Type	Required	Description
`meetingId`	`string`	Yes	—

Answers structuredNotes, markdownContent and topics. A note carries a segmentId, a timestamp in seconds and a topicId that matches a topic’s id. Output

Name	Type	Required	Description
`structuredNotes`	`object[]`	Yes	—
`markdownContent`	`string`	Yes	The notes rendered as markdown
`topics`	`object[]`	Yes	—

structuredNotes full type

{
  segmentId: string,
  timestamp: number,
  text: string,
  topicId: string
}[]

topics full type

{
  id: string,
  order: number,
  title: string,
  summary: string
}[]

transcript

meetings.transcript Get the transcript of one meeting, sentence by sentence Risk: read

await corsair.tldv.api.meetings.transcript({ meetingId: 'meeting-123' });

Input

Name	Type	Required	Description
`meetingId`	`string`	Yes	—

Each entry in data is one sentence with a speaker, text, startTime and endTime, with the offsets in seconds. This is the same object a TranscriptReady webhook carries. Output

Name	Type	Required	Description
`id`	`string`	Yes	—
`meetingId`	`string`	Yes	—
`data`	`object[]`	Yes	—

data full type

{
  speaker: string,
  text: string,
  startTime: number,
  endTime: number
}[]

Overview

Plugins

Folders

list

Meetings

get

import

list

notes

transcript

​Folders

​list

​Meetings

​get

​import

​list

​notes

​transcript

Folders

list

Meetings

get

import

list

notes

transcript