Niches · Meeting capture & meeting infrastructure

Answer a question from the user's own recorded meetings

Two or three companies' docs show a full fit.

2companies, full fit
4companies, fit with a limit
15companies carry the tag
13share of the category's work
9first-hand accounts

What counts as done

The right meeting found by participant, date or text in one search call, and its full transcript with named speakers returned across as many pages as needed, with a cursor that says when the end is reached, plus the meeting link to cite.

How agents ask for it: what did we decide about pricing last week, pull the transcript of my call with Acme, what did the customer ask for, summarise my meetings on project X

Does the job 2

The docs show it meets the test above, and nothing we read says otherwise.

ListingWhy
Circleback B
Circleback AI, Inc.
Search meetings takes searchTerm over names and content and attendeeProfileIds, list meetings filters by attendee, domain and date, Get transcript returns the full transcript with speaker names, and lists page by Link rel=next. Free keeps only 30 days. source
Fireflies.ai B
Fireflies.AI Corp.
transcripts takes keyword with scope all (titles and sentences), participants, organizers and a date range, up to 50 a page, and transcript returns every sentence with speaker names; MCP tools mirror it. acc-14 pulled a transcript this way into Obsidian. source

Does the job, with a limit 4

It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.

ListingWhy
Fathom C
Fathom Video Inc.
GET /meetings filters by invitee domain, recorder, team and created dates and returns share_url and transcripts with speaker display_name, but pages are fixed at 10, there's no attendee-name or text search, and the MCP tools aren't published. source
Granola B
Granola, Inc.
Search notes filters by participant, date and keywords, transcript search finds what was said, Get transcript pages with cursor and hasMore, and notes carry web_url. Speakers are often only me or them (acc-45), Basic withholds transcript tools, and acc-44 saw MCP transcripts cut at page one. source
MeetGeek C
AIVISION PRODUCTS SRL
Transcripts page by limit up to 500 and next_cursor with speaker names, but GET /v1/meetings filters by nothing beyond limit and cursor and returns only ids and times, so finding a meeting by participant or text takes many calls against 100 requests a day on Free and Pro. source
tl;dv D
tldx Solutions GmbH
GET /meetings takes query, from, to, meetingType and onlyParticipated, up to 100 a page, and returns url and invitees; the transcript has speaker per sentence. There's no participant filter, and a meeting is readable only if its organiser is on Pro or Business. source

Couldn't check 1

The docs were closed to us, or silent. These carry no mark either way.

ListingWhy
Read AI E
Read AI, Inc.
Search filters, transcript paging and speaker labels are in the REST and MCP reference, which returned 403; acc-49 is a connector plan, not a result.

Doesn't do this job 6

Listings in the category that do something else, with the reason.

ListingWhy
Attendee D
Attendee Labs, Inc.
Bot infrastructure, not a notetaker library: List Bots filters only by state, meeting_url, deduplication key and join time, with no participant or transcript-text search. source
Meeting BaaS B
SAS SPOKE
Bot infrastructure, not a person's notetaker library: List Bots filters by bot or participant name, dates, platform and status, with no transcript-text search. source
MeetStream C
MeetStream Corp.
Bot infrastructure: List Bots filters by status, date, platform and custom attributes, with no participant or transcript-text search over a person's library. source
Nylas Notetaker API BB
Nylas, Inc.
A bot API for the developer's own Notetakers, not a person's searchable meeting library; no participant or transcript-text search is documented. source
Recall.ai B
Hyperdoc Inc.
Infrastructure for the developer's own bots: List Bots filters by join time, meeting URL, platform and status, with no participant or transcript-text search over a person's library. source
Zoom Realtime Media Streams (RTMS) C
Zoom Communications, Inc.
A live stream, not a meeting library; there's no search or stored transcript to query. source

What stops agents finishing

  • transcripts withheld on the free plan (Granola Basic) or when the organiser is on Free (tl;dv)
  • MCP search covers titles and summaries, not transcript text (Fathom, per a vendor-comparison page)
  • long transcript silently truncated at the first page (acc-44)
  • speakers labelled Me and Them (acc-45)
  • Read AI returns nothing until an admin enables downloads

Around it

Before: chat.user-question. After: productivity.write-doc, project-management.capture-tasks, company-knowledge.cited-answer.

Where this comes from

Weight 13 of 100 in Meeting capture & meeting infrastructure: 9 of 56 accounts (acc-13, acc-14, acc-15, acc-17, acc-21, acc-44, acc-45, acc-46, acc-47); the Granola, Fathom and Fireflies MCP accounts

The same as data: /api/v1/use-cases/meeting-capture.ask-my-meetings.json, or the MCP tool get_use_case with id meeting-capture.ask-my-meetings. Other jobs in this category: /api/v1/niches?category=meeting-capture.

Draft, read 2026-11-15 from public docs and first-hand accounts, not yet reviewed by a person. A mark says what the docs show, not what a run measured.

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.