ToolsConnector
Granola
Read Granola meeting notes, attendees, AI summaries and full transcripts, and get told of new notes, from Claude or ChatGPT through Tulina's MCP server.
Official site: granola.ai
What it can do
3 read/1 write
- 01Read
List meeting notes by date window or folder
- 02Read
Read a call's title, date, participants and AI summary
- 03Read
Read the full transcript of a call, page by page
- 04Write
Create and manage webhook endpoints that fire on new notes
A call is recorded by whoever hosted it, not by every attendee, so the person you are looking for often sits on a colleague's call rather than their own. Matching the company domain as well as the exact address is what separates a hit from a false negative here.
Does Granola AI have an MCP?
Yes. Granola runs its own MCP server at https://mcp.granola.ai/mcp,
signed in through the browser with OAuth, with tools to query meetings
in plain language, list and read notes, read transcripts and folders.
Tulina reaches Granola differently, through Granola's public API with a
key held in the vault, which is what lets a hosted agent read a call
with nobody signed in.
Does Granola have an API?
Yes, a REST API with personal keys for members of a Business plan and workspace keys that an admin creates on Enterprise. It serves notes, transcripts, folders and webhook endpoints, and it is what this connector is built on. Granola applies its own rate limits per key.
How do you connect Granola to Claude and ChatGPT?
Through Tulina's own MCP server, not Granola's own hosted one.
- In the Granola desktop app, open Settings, then Connectors, then API keys, and create a key. A workspace key covers the notes the workspace shares; a personal key covers that person's notes.
- Paste it into Tulina's Granola connector, for yourself or for the organisation.
- Add
https://mcp.tulina.ai/mcpto Claude or ChatGPT and sign in. The MCP page has the link.
Then list recent meeting notes, open a call to read its title, date, attendees and AI summary, or read the transcript, in the same chat as the account research or deal prep it is for.
Tulina vs Granola's official MCP server
| Granola's MCP server | Tulina | |
|---|---|---|
| Endpoint | mcp.granola.ai/mcp, hosted by Granola | mcp.tulina.ai/mcp, one link for every connected tool |
| Sign-in | Browser OAuth, or enterprise-managed authorisation. No API key or service account | A Granola API key held in Tulina's vault |
| Runs with nobody signed in | Works through the OAuth session a person opened | Hosted agents run on a schedule or a webhook with the stored key |
| Plain-language search across meetings | Yes, its query tool | No search endpoint; notes are listed by date or folder and opened |
| Free plan | Personal notes from the last 30 days; transcript and folder tools on paid plans | Needs an API key, which Granola issues on Business and Enterprise plans |
| New-note webhooks | Not part of its tool list | Create, update and delete webhook endpoints |
For one person asking about their own meetings, Granola's server is the quicker start. Tulina fits when a call feeds a ticket, a CRM note or a follow-up email, and when that has to happen without someone at a keyboard.
What you can ask it
"What did we agree with Acme on Tuesday's call?" The agent lists the
notes created that day, opens the ones whose attendees carry an
acme.com address, and answers from the summary. Because a call lives
with its host, a colleague's call is checked as well as your own.
"Pull the exact words the customer used about pricing." The summary paraphrases. The agent reads the transcript of that call page by page and quotes the passage. A long meeting's transcript is paginated, so nothing is truncated without saying so.
"Every new call with a customer, send the notes to the team channel." The agent creates a Granola webhook endpoint for the workspace's notes. Each new note then reaches a hosted agent, which reads it and posts the summary to Slack. The signing secret is returned once, at creation.
Pull insights from call recordings, Prepare for a customer call and Call feedback becomes tickets and docs all start from Granola. For teams on another recorder, the Fireflies and Grain connectors cover the same ground.
What the Granola connector can't do
- Search by person, company or keyword. Granola's API has no search endpoint, so finding a call means listing a window of dates or a folder and reading the matching notes. Keep the window narrow.
- Read notes the key cannot see. A personal key reaches that person's notes and those shared with them; the rest of the workspace needs a workspace key.
- Read Granola's audit log. The endpoint answered not found on the key it was tested with, so it is left out.
- Edit, delete or write notes. Webhook endpoints are the only thing it creates.
- Read screenshots or attachments from a meeting.
Why through Tulina
The connection was never the hard part.
Most systems now publish their own way for a model to reach them, and that is good: it is what Tulina connects to. The hard part is deciding who may use it, what it is allowed to return, and what the model knows before it asks.
The keys stay in one place
YouTeamOrganisation
Connect Granola once. The credential is encrypted in a vault, never held in clear, and never pasted into a chat or a workflow file. A person uses their own key where they have one, their team's where they do not, the organisation's after that, so nobody is blocked and nobody holds more access than they need. Revoke it once and everything running on it stops.
What leaves is filtered before it leaves
fields droppedmasked or hashedpseudonymised
Roles decide who can reach Granola at all. Beyond that, a policy on the connector itself decides what comes back. The data is minimised at the source rather than after it has already reached a model.
Every call is on the record
whatwhowhen
Runs are logged for the organisation. That is what makes an automated action reviewable after the fact, and it is what a security review asks for first.
The run starts knowing your company
namingkey accountsdecisions
A direct connection gives a model access. It does not tell it how your company names things, which accounts matter, or what was decided last quarter. Tulina carries that across every tool, so the same request against Granola returns something usable instead of something generic.
Processes that run on Granola
Call feedback becomes tickets and docs
Every call recorded in the day is distilled overnight, with bugs and feature requests routed to the right team as tickets and docs, plus a Friday digest.
5 steps →Expand accounts from usage
Each morning, accounts near a plan limit or on a feature their plan lacks get an offer priced on your packaging rules, an expansion deal and a drafted proposal.
7 steps →Follow champions who change jobs
Each week, champions at won accounts are checked for a new employer, the new company is scored on your ICP, and a warm deal opens with a note for the right rep.
7 steps →Follow through on sales calls
After every sales call, the follow-up email is drafted from what was said, the deal gets a dated next step, product asks are filed and any risk is flagged.
7 steps →Prepare for a customer call
Before the call, the last conversations, the account's state, how the product is actually used, and where requested features stand, in one brief.
5 steps →Prepare for a demo from the last call
The last calls, the deal's state and fresh research become one prep brief, plus a deck built on that company's own use cases, posted before the demo.
5 steps →Pull insights from call recordings
Conversation intelligence from a name, company or email: the right sales call found in your CRM or recordings, returned with a summary, the link saved back.
5 steps →
