mcp-granola

Retrieve Granola meeting transcripts and notes via code_execution callbacks.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/SillyHippy/PAULFAM --skill mcp-granola
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mcp-granola
Source: https://github.com/SillyHippy/PAULFAM/tree/main/.local/mcp_skills/granola
Command: npx skills add https://github.com/SillyHippy/PAULFAM --skill mcp-granola

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides programmatic access to Granola MCP meeting data and transcripts via code_execution callbacks, enabling AI agents to fetch exact content and meeting context from Granola.

Core Features & Use Cases

  • Four pre-registered functions for Granola data: mcpGranola_getMeetingTranscript, mcpGranola_getMeetings, mcpGranola_listMeetings, mcpGranola_queryGranolaMeetings.
  • Use cases include retrieving verbatim transcripts, gathering meeting metadata and AI-generated summaries, and performing natural-language queries over Granola notes with inline citations.

Quick Start

Ask the agent to retrieve a meeting's full transcript by providing its meeting_id and calling mcpGranola_getMeetingTranscript.

Frequently Asked Questions about mcp-granola

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I access Granola meeting transcripts programmatically?

Retrieve Granola meeting transcripts programmatically by invoking the mcpGranola_getMeetingTranscript callback with a specific meeting_id via code_execution. This fetches the verbatim transcript content directly for AI agent processing.

Can I query Granola notes using natural language?

Query Granola notes using natural language by calling the mcpGranola_queryGranolaMeetings function. This supports natural-language queries over meeting data and returns results with inline citations for contextual accuracy.

What functions are available for retrieving Granola meeting data?

Four pre-registered functions are available for retrieving Granola meeting data: mcpGranola_getMeetingTranscript, mcpGranola_getMeetings, mcpGranola_listMeetings, and mcpGranola_queryGranolaMeetings. These cover transcript retrieval, metadata gathering, and meeting listing.

How do I list all available Granola meetings?

List all available Granola meetings by executing the mcpGranola_listMeetings callback through code_execution. This function returns available meeting entries, allowing AI agents to enumerate and select specific meetings for further data retrieval.

Do I need pre-registered callbacks to access Granola MCP data?

Yes, accessing Granola MCP data requires pre-registered callbacks. The Skill depends on the presence of mcpGranola_getMeetingTranscript, mcpGranola_getMeetings, mcpGranola_listMeetings, and mcpGranola_queryGranolaMeetings to fetch meeting content.

Can I get AI-generated summaries from Granola meetings?

Gather AI-generated summaries from Granola meetings by using the mcpGranola_getMeetings function. This callback retrieves meeting metadata and AI-generated summaries alongside the core meeting context via code_execution.