# Sync or back-fill a whole meeting library into a knowledge base

> Sync or back-fill a whole meeting library into a knowledge base: 0 companies fully do it, 11 with a limit, 15 carry the tag. Every listing in Meeting capture & meeting infrastructure with the reason and a source.

- Canonical: https://www.anchorterminal.com/niches/meeting-capture.library-sync/
- Markdown: https://www.anchorterminal.com/niches/meeting-capture.library-sync/index.md (~1,350 tokens)
- Slim: https://www.anchorterminal.com/niches/meeting-capture.library-sync/index.min.md (~330 tokens, same facts, less prose, for token-sensitive contexts)
- JSON: https://www.anchorterminal.com/niches/meeting-capture.library-sync/index.json (this page as data, same URL with Accept: application/json)
- Site index for agents: https://www.anchorterminal.com/llms.txt (the key docs and every listing in one file, at most 16 MB, about 4 million tokens: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- MCP: https://www.anchorterminal.com/mcp (the whole catalogue as tools, no key)
- DNS: dig +short TXT <slug>.grade.anchorterminal.com (the grade as a DNS TXT record, for a sandbox with no HTTP where names still resolve; also top., best.<category>. and <a>.vs.<b>. under the zone, and the same under grade.letme.dev)
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)
- Updated: 2026-11-15

No company's docs show a full fit yet. The listings below do the job with a limit, or not at all.

| Companies, full fit | With a limit | Carry the tag | Share of the category's work | First-hand accounts |
| --- | --- | --- | --- | --- |
| 0 | 11 | 15 | 9 | 6 |

## What counts as done

Every meeting in a date window listed through a cursor without gaps, with transcripts fetched within the documented rate limit, an updated-since filter or event for incremental runs, and a deletion signal so removed meetings can be forgotten.

Measured, for our own tests: 500 meetings listed and fetched without a 429 that the client cannot recover from.

How agents ask for it: import all my Fathom calls; index our meeting notes for search; keep our knowledge base in step with Granola; archive every meeting

## Does the job, with a limit

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.

- **Attendee (Attendee Labs, Inc.)**: List Bots pages by cursor with join_at window filters and data_deleted is a state, but there's no updated-since filter on bots, no deletion event, and no hosted rate limit is documented.
- **Circleback (Circleback AI, Inc.)**: List meetings pages by cursor with createdAtGte and createdAtLte, but there's no updated-since filter or deletion event, Free allows 20 requests a minute and 30 days of history.
- **Fathom (Fathom Video Inc.)**: GET /meetings pages by next_cursor with created_after and Retry-After on 429, but 10 a page can't be raised, heavy calls are 30 a minute and as few as 5, and there's no updated-since filter or deletion signal.
- **Fireflies.ai (Fireflies.AI Corp.)**: transcripts pages by limit and skip offset over fromDate and toDate, but there's no updated-since filter or deletion event, and Free caps at 500 requests a day and Pro at 5,000, so 500 meetings plus transcripts exceeds Free.
- **Granola (Granola, Inc.)**: List notes pages by cursor up to 30 a page with created and updated_after filters at 5 requests a second, but deleted_at is visible only to keys with access to all workspace notes, and the API needs Business.
- **MeetGeek (AIVISION PRODUCTS SRL)**: GET /v1/meetings pages by cursor up to 500 a page, but there's no date window, updated-since filter or deletion event, and Free and Pro stop at 100 requests a day, far short of 500 transcripts.
- **Meeting BaaS (SAS SPOKE)**: List Bots pages by cursor with limit, created_after and ended_after, and GETs don't count against the rate limit, but there's no updated-since filter or deletion event, and plan retention of 3 to 30 days caps any back-fill.
- **MeetStream (MeetStream Corp.)**: List Bots pages with nextCursor and hasNextPage over a from and to date window, and a data_deletion webhook exists, but there's no updated-since filter and read rate limits aren't stated in what we read.
- **Nylas Notetaker API (Nylas, Inc.)**: Notetakers can be listed and notetaker.deleted signals removals, but media is deleted after 14 days by default (30 at most), no updated-since filter was found, and no Notetaker rate limit is documented.
- **Recall.ai (Hyperdoc Inc.)**: List Bots pages by cursor within a join_at window at 60 requests a minute with Retry-After, and recording.deleted webhooks signal deletions, but there's no updated-since filter.
- **tl;dv (tldx Solutions GmbH)**: GET /meetings pages by page number up to 100 within a from and to window, capped at 10,000 results, but there's no cursor, updated-since filter, deletion event or published rate limit.

## Couldn't check

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

- **Read AI (Read AI, Inc.)**: List paging, filters and rate limits are in the API reference, which returned 403.

- not this job: Zoom Realtime Media Streams (RTMS) (Zoom Communications, Inc.): Nothing is stored for later listing; RTMS sends data only while a session runs.

## What stops agents finishing

- 10 meetings a page with no size parameter (Fathom, Read AI)
- daily request caps of 100 (MeetGeek Free and Pro) or 500 (Fireflies Free)
- no updated-since filter or deletion event
- model credits run out mid-batch (acc-52)

## Around it

After: agent-memory.ingest-records, company-knowledge.cited-answer, file-storage.backup-archive.

## Where this comes from

Weight 9 of 100 in Meeting capture & meeting infrastructure: 6 of 56 accounts (acc-44, acc-46, acc-48, acc-49, acc-52, acc-56); three are connector proposals rather than finished work

As data: https://www.anchorterminal.com/api/v1/use-cases/meeting-capture.library-sync.json, or the MCP tool get_use_case with id meeting-capture.library-sync. Draft, read 2026-11-15 from public docs and first-hand accounts, not yet reviewed by a person.

---

Query this catalogue directly over MCP: https://www.anchorterminal.com/mcp (Streamable HTTP, no key). How: https://www.anchorterminal.com/agents/#mcp
