Sign in Book a demo
Back to Tools
Productivity & Docs API Key BYOC
Build Your Agent View Granola Documentation
granola logo

Granola MCP Tools for AI Agents

Let an agent list meetings, inspect meeting metadata, and retrieve timestamped transcripts from Granola.

Available Tools

3 tools available in this integration

Granola.GetMeeting

Get full metadata for one Granola meeting by its note ID.

Granola.GetMeetingTranscript

Get the verbatim transcript for a Granola meeting, with speaker labels and timestamps.

Granola.ListMeetings

List Granola meetings, optionally filtered by date range, with cursor pagination.

Why agent-optimized tools matter

Built for intent, not API schemas

Ask the agent to "find last week's customer meetings and pull the transcript for the one about renewal." Arcade helps it use the right Granola actions without guessing meeting note IDs, date filters, or pagination details.

Up to 100x fewer tokens

Granola meeting lists return lightweight metadata, and full transcripts are retrieved only when needed. The agent can narrow the meeting set first instead of loading every transcript into the model.

Complete the workflow end to end

List meetings, narrow by date, open the matching meeting, read its transcript, and return speaker-attributed findings through one toolkit designed around the complete meeting-intelligence task.

Make actions predictable and controlled

Arcade tools return structured results and useful errors across users and runs. Teams can apply scope restrictions, redaction, and approval policies so agents follow the same controls every time.

Securing Granola

The more capabilities agents get, the more useful they are. However, more capabilities bring more risks of undesired effects such as data loss and exfiltration.

This is why Arcade provides you with contextual rules that dictate how and when a tool is allowed to run. Arcade’s contextual access lets you govern tool visibility and behavior at every stage of execution and control who can see a tool, what inputs are allowed, and what comes back.

Learn about Contextual Access Policies

For Granola, you can implement some of these policies:

Access when tools are listed for a user
  • Decide which tools an agent is allowed to see at all. A tool that is never listed cannot be called, whoever is asking.
Pre-execution before each tool call
  • Inspect the inputs a call is about to run with, and block it or send it for human approval when they fall outside what the agent should do unsupervised.
Post-execution after the tool returns
  • Redact attendee email addresses, private meeting links, and location details from Granola.GetMeeting before they reach the model, so an agent cannot expose participant contact data.
  • Strip phone numbers, access codes, and customer identifiers from Granola.GetMeetingTranscript results before the model sees the transcript, so private details are not repeated in generated notes or replies.
  • Redact meeting titles, attendee names, and calendar metadata from Granola.ListMeetings results unless the requesting agent is working in an approved project scope, so an inbox-wide scan cannot reveal confidential meeting activity.

How to use Granola with your agent

Pick your MCP client below and follow the setup to connect Granola to your agent. Configure the shared API key once, then retrieve the meeting data your agent needs.

Your questions, answered

A Granola MCP server exposes Granola meeting data as structured tools an AI agent can call through the Model Context Protocol. With Arcade, an agent can list meetings, retrieve meeting metadata, and read transcripts with speaker labels and timestamps.

Granola's public Enterprise API is not an MCP server. Arcade provides an MCP and tool runtime around it, including authenticated access, agent-optimized meeting actions, approval workflows, redaction, and controls that work across supported clients.

Arcade agents can list Granola meetings by date range, retrieve full metadata for a meeting by note ID, and read its verbatim transcript with speaker labels and timestamps. These actions are read-only; they do not create, edit, or delete meetings.

Arcade uses the shared Granola Enterprise API key or secret configured for the integration. The credential determines what the API can access; this toolkit's available actions only read meeting metadata and transcripts.

No. Granola uses one shared API key or secret configured once for the integration, rather than a separate Granola connection for every user. Arcade brokers the credential instead of passing it to the model.

Yes. Contextual Access policies can require human approval before a Granola call that your organization considers consequential, and the call proceeds only after the policy is satisfied. The available Granola actions are read-only, so there are no write or destructive actions to approve.

Arcade proxies Granola API calls and does not persist your meeting content as part of normal tool execution. The Granola credential is brokered and stored encrypted at rest, and the model never receives it.

Any framework or client that can connect to Arcade's MCP gateway or tool APIs works with Arcade Granola, including LangChain, OpenAI Agents, CrewAI, Claude Desktop, Claude Code, Cursor, Visual Studio Code, Microsoft Copilot Studio, and GitHub Copilot.

Arcade maintains access while the configured Granola credential remains valid. If the key is revoked, rotated, expires, or can no longer be used, calls stop until the credential is reconfigured.

Granola's Enterprise API rate limits and quotas still apply to calls made through Arcade. Use the limits associated with your Granola API access for exact thresholds; the available actions do not bypass those limits.

Ready to actually ship your agent?

Connect your agent to Granola with a shared API key, agent-optimized tools and approval controls for consequential actions.

Build Your Agent Read the Granola Quickstart