granola-ai-mcp-server

nehiljain/granola-ai-mcp-server

3.3

If you are the rightful owner of granola-ai-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.

Granola MCP Server is an experimental Model Context Protocol server designed to integrate Granola.ai meeting intelligence with Claude Desktop, leveraging local cache for data processing.

Tools
5
Resources
0
Prompts
0

Granola MCP Server (Experimental)

An experimental Model Context Protocol (MCP) server for integrating Granola.ai meeting intelligence with Claude Desktop. This uses Granola's local cache. I don't have any idea how Granola updates or maintains that cache. My guess is that it's doing a rolling context window, but storing long term notes up in AWS. So, YMMV. Use at your own risk. I will likely add a cache shipper at some point since we don't have access to Gronala's data in the cloud.

Features

  • Meeting Search: Search meetings by title, content, participants, and transcript content
  • Meeting Details: Get comprehensive meeting metadata with local timezone display
  • Full Transcript Access: Retrieve complete meeting conversations with speaker identification
  • Rich Document Content: Access actual meeting notes, summaries, and structured content
  • Meeting Export: Export meetings to markdown files with auto-generated filenames
  • Pattern Analysis: Analyze patterns across meetings (participants, frequency, topics)
  • Timezone Intelligence: All timestamps automatically display in your local timezone
  • Real-time Integration: Seamlessly connects to your actual Granola meeting data

Quick Start

Prerequisites

  • Python 3.12+
  • uv package manager
  • macOS with Granola.ai installed
  • Claude Desktop application
  • Granola cache file at ~/Library/Application Support/Granola/cache-v3.json

Installation

  1. Clone and navigate to the project:

    git clone <repository-url>
cd granola-ai-mcp

  2. Install dependencies with uv:

    uv sync

  3. Test the installation:

    uv run granola list

  4. Configure Claude Desktop by adding to your claude_desktop_config.json:

    {
  "mcpServers": {
    "granola": {
      "command": "uv",
      "args": ["--directory", "/absolute/path/to/granola-ai-mcp", "run", "granola-mcp-server"],
      "env": {}
    }
  }
}

    Important: Replace /absolute/path/to/granola-ai-mcp with your actual project path.

  5. Restart Claude Desktop to load the MCP server

Global Installation & Upgrading

To install the Granola CLI globally so you can run granola commands from anywhere:

# Install globally with uv (from anywhere)
uv tool install granola-ai-mcp-server

# Or install from local source
uv tool install --editable /path/to/granola-ai-mcp-server

To upgrade to the latest version:

# Upgrade to latest version
uv tool install --upgrade granola-ai-mcp-server

# Or upgrade from local source
uv tool install --upgrade --editable /path/to/granola-ai-mcp-server

After installing/upgrading, you can run granola commands from anywhere:

# These commands now work from any directory
granola list
granola search "meeting topic"
granola get
granola export --output-dir ./meetings/
granola clean

Note: For the MCP server integration with Claude Desktop, you'll still need to use the local project path in your claude_desktop_config.json:

{
  "mcpServers": {
    "granola": {
      "command": "uv",
      "args": ["--directory", "/absolute/path/to/granola-ai-mcp-server", "run", "granola-mcp-server"],
      "env": {}
    }
  }
}

CLI Tool

The Granola CLI provides powerful command-line access to your meeting data with rich formatting and interactive features.

Installation & Setup

The CLI is automatically available after installing the project:

# Install dependencies
uv sync

# Test CLI is working
uv run granola --help

Available Commands

granola list - List All Meetings

Display meetings in a beautiful formatted table with key information.

# List all meetings
uv run granola list

# Show ignored meetings
uv run granola list --show-ignored

# Use custom cache path
uv run granola --cache-path /path/to/cache.json list

# Set timezone for display
uv run granola --timezone America/Los_Angeles list

Output includes:

  • Meeting title, date, duration, platform
  • Transcript statistics (word count)
  • Participant count
  • Meeting ID
granola get - Get Meeting Details

Retrieve comprehensive meeting information including full transcripts.

# Interactive meeting selection
uv run granola get

# Get specific meeting by ID
uv run granola get c96cbb7e-b0ab-40e8-bbad-bd4efc449af3

# Copy output to clipboard
uv run granola get --clipboard

# Save to file
uv run granola get --output meeting_notes.md

# Combine options
uv run granola get --clipboard --output backup.md

Output includes:

  • Complete meeting metadata (ID, title, date, duration, platform)
  • Organizer and creator information
  • Participant list
  • Full transcript with speaker identification
  • Meeting notes and documents
  • Transcript statistics (word count, duration)
granola search - Fuzzy Search Meetings

Search meetings using fuzzy matching on titles, attendees, or transcripts with multiple output formats.

# Basic title search (default behavior)
uv run granola search "project planning"

# Search participant names
uv run granola search "john" --attendees

# Search transcript content (slower but comprehensive)
uv run granola search "budget discussion" --transcript

# Combine search types
uv run granola search "meeting" --attendees --transcript

# Include ignored meetings
uv run granola search "test" --show-ignored

# JSON output for programmatic use
uv run granola search "quarterly review" --json

# YAML output for configuration files
uv run granola search "team sync" --yaml

# Include meetings with no transcript content
uv run granola search "standup" --include-empty-transcripts

# Complex search with multiple options
uv run granola search "architecture" --attendees --transcript --json --include-empty-transcripts

Search Features:

  • Fuzzy Matching: Uses intelligent fuzzy matching (60%+ threshold)
  • Multiple Fields: Search titles, attendees, or transcript content
  • Smart Filtering: By default excludes meetings with no transcript content
  • Flexible Output: Table (default), JSON, or YAML formats
  • Match Scoring: Shows what matched and confidence score
  • Date Sorting: Results sorted by date (most recent first)

Output Formats:

  • Table: Compact table view with key information
  • JSON: Structured data for scripts and APIs
  • YAML: Human-readable format for configuration
granola clean - Clean Up Meetings

Identify and manage meetings with minimal content or missing transcripts.

# Dry run (default) - see what would be cleaned
uv run granola clean

# Actually clean meetings
uv run granola clean --confirm

# Custom word threshold
uv run granola clean --min-words 1000

# Duration threshold
uv run granola clean --min-duration 30

# Reset ignore list
uv run granola clean --reset

# Combine criteria
uv run granola clean --min-words 500 --min-duration 15 --confirm

Cleaning criteria:

  • Meetings with no transcript
  • Meetings with transcript below word threshold
  • Meetings below duration threshold
  • All criteria can be combined
granola export - Export Meeting to Markdown File

Export a meeting to a formatted markdown file with auto-generated filename.

# Interactive meeting selection
uv run granola export --output-dir ./meetings/

# Export specific meeting by ID
uv run granola export c96cbb7e-b0ab-40e8-bbad-bd4efc449af3 --output-dir ~/Documents/meetings/

# Short flag
uv run granola export -o ./output/

# Export to home directory
uv run granola export --output-dir ~/meetings/

Features:

  • Auto-generated filename: Files are named YYYY-MM-DD-meeting-title.md
  • Three sections: Metadata, Notes, and Transcript
  • Overwrite protection: Asks for confirmation if file exists
  • Directory creation: Prompts to create output directory if it doesn't exist
  • Validation: Loudly errors if meeting has no notes content

Output format:

# Meeting Title

## Metadata
- **ID:** meeting-id
- **Title:** Meeting Title
- **Date:** 2025-10-09 14:30
- **Duration:** 60 minutes
- **Platform:** Zoom
- **Participants:**
  - John Doe
  - Sarah Smith

## Notes
[Meeting notes and documents from Granola]

## Transcript
**Speakers:** John Doe, Sarah Smith
**Word Count:** 2,500
[Full transcript content]

Filename examples:

  • 2025-10-09-quarterly-planning-sync.md
  • 2025-10-08-weekly-team-standup.md
  • 2025-10-07-architecture-review.md

CLI Usage Examples

Basic Operations
# Quick meeting overview
uv run granola list

# Search for specific meetings
uv run granola search "project planning"

# Get details for most recent meeting
uv run granola get

# Find meetings to clean up
uv run granola clean
Advanced Usage
# Export meeting with auto-generated filename
uv run granola export --output-dir ./meetings/

# Export specific meeting to directory
uv run granola export abc123 -o ~/Documents/meetings/

# Copy meeting details to clipboard for sharing
uv run granola get --clipboard

# Clean up meetings with less than 100 words
uv run granola clean --min-words 100 --confirm

# Show only ignored meetings
uv run granola list --show-ignored
Search Examples
# Find meetings by title
uv run granola search "quarterly review"
uv run granola search "standup"
uv run granola search "architecture"

# Search by participant names
uv run granola search "john" --attendees
uv run granola search "sarah" --attendees
uv run granola search "team lead" --attendees

# Search transcript content
uv run granola search "budget" --transcript
uv run granola search "deadline" --transcript
uv run granola search "technical debt" --transcript

# Combined searches
uv run granola search "planning" --attendees --transcript
uv run granola search "review" --attendees --transcript --json

# Include empty transcripts
uv run granola search "meeting" --include-empty-transcripts
uv run granola search "sync" --include-empty-transcripts --yaml

# Programmatic output
uv run granola search "project" --json > search_results.json
uv run granola search "team" --yaml > team_meetings.yaml
Batch Operations
# Clean up short meetings (less than 15 minutes)
uv run granola clean --min-duration 15 --confirm

# Reset all ignored meetings
uv run granola clean --reset

# Export multiple meetings (run multiple times)
uv run granola get --output meeting1.md
uv run granola get --output meeting2.md

Quick CLI Workflows

Daily Meeting Review
# See today's meetings
uv run granola list

# Search for specific topics
uv run granola search "project planning"
uv run granola search "john" --attendees

# Get details of most recent meeting
uv run granola get

# Export important meeting to file
uv run granola get --output important_meeting.md
Meeting Search & Discovery
# Find meetings by topic
uv run granola search "architecture" --transcript
uv run granola search "budget" --attendees --transcript

# Search for specific people
uv run granola search "sarah" --attendees
uv run granola search "team lead" --attendees --json

# Find meetings with specific content
uv run granola search "deadline" --transcript --yaml
uv run granola search "technical debt" --transcript --include-empty-transcripts

# Export search results
uv run granola search "quarterly review" --json > q4_reviews.json
uv run granola search "standup" --yaml > daily_standups.yaml
Meeting Cleanup
# See what can be cleaned
uv run granola clean

# Clean up meetings with no transcripts
uv run granola clean --confirm

# Clean up short meetings
uv run granola clean --min-duration 10 --confirm
Meeting Export
# Export meeting with auto-generated filename
uv run granola export --output-dir ./meetings/

# Export specific meeting
uv run granola export abc123 -o ~/Documents/meetings/

# Copy meeting to clipboard for sharing
uv run granola get --clipboard

# Export multiple meetings to a directory
uv run granola export meeting_id_1 -o ./archive/
uv run granola export meeting_id_2 -o ./archive/

# Export with manual filename (using get)
uv run granola get --output custom_name.md

CLI Features

Interactive Selection

When using granola get without a meeting ID, you'll see an interactive menu:

? Select a meeting: (Use arrow keys)
 » ⏸ Nehil / Brent | 2025-10-09 15:30 | 1 participants
   Ray POC on AWS Questions from Fahri and Nehil and Soma | 2025-10-09 14:01 | 3 participants
   Anyscale platform demo for PayPal machine learning infrastructure | 2025-10-09 13:30 | 0 participants
Rich Output Formatting
  • Tables: Beautiful formatted tables with colors and alignment
  • Markdown: Structured output with headers, lists, and formatting
  • Statistics: Word counts, character counts, duration information
  • Timezone: Automatic local timezone display
  • Search Results: Multiple output formats (table, JSON, YAML)
  • Fuzzy Matching: Intelligent search with confidence scores
Clipboard Integration
  • macOS: Uses pbcopy for clipboard functionality
  • Fallback: Graceful fallback if clipboard tools unavailable
  • Confirmation: Clear feedback when content is copied
File Output
  • Markdown format: Structured, readable meeting notes
  • UTF-8 encoding: Proper character encoding for all content
  • Large files: Handles transcripts with 25,000+ characters
Search Output Examples

Table Output (Default):

                           Search Results for 'project planning'                           
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━┳━━━━━━━━━━┳━━━━━━━━━━┳━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Title                                     ┃ Date     ┃ Dur ┃ Platform ┃ Transcr… ┃ Ppl ┃ Match     ┃ ID       ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━╇━━━━━━━━━━╇━━━━━━━━━━╇━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━┩
│ Q4 Planning Meeting                       │ 10/09    │ 60m │ Zoom     │ 2.5k     │ 5   │ Tit (95%) │ abc123…  │
│ Project Kickoff Discussion               │ 10/08    │ 45m │ Teams    │ 1.8k     │ 3   │ Tit (87%) │ def456…  │
└───────────────────────────────────────────┴──────────┴─────┴──────────┴──────────┴─────┴───────────┴──────────┘

JSON Output:

{
  "query": "project planning",
  "total_results": 2,
  "meetings": [
    {
      "id": "abc123...",
      "title": "Q4 Planning Meeting",
      "date": "2025-10-09 14:30",
      "duration_minutes": 60,
      "platform": "Zoom",
      "participants": ["john", "sarah", "mike"],
      "participant_count": 3,
      "transcript_stats": "2,500 words",
      "match": {
        "type": "Title",
        "score": 95
      }
    }
  ],
  "search_options": {
    "attendees": false,
    "transcript": false,
    "show_ignored": false,
    "include_empty_transcripts": false
  }
}

YAML Output:

query: project planning
total_results: 2
meetings:
- id: abc123...
  title: Q4 Planning Meeting
  date: 2025-10-09 14:30
  duration_minutes: 60
  platform: Zoom
  participants:
  - john
  - sarah
  - mike
  participant_count: 3
  transcript_stats: 2,500 words
  match:
    type: Title
    score: 95
search_options:
  attendees: false
  transcript: false
  show_ignored: false
  include_empty_transcripts: false

CLI Configuration

Environment Variables
# Custom cache path
export GRANOLA_CACHE_PATH="/custom/path/cache.json"

# Custom timezone
export GRANOLA_TIMEZONE="America/Los_Angeles"
Command Line Options
# Global options (available for all commands)
--cache-path TEXT    Path to Granola cache file
--timezone TEXT      Timezone for date display
--help               Show help message

# List command options
--show-ignored       Show ignored meetings

# Get command options
--clipboard, -c      Copy output to clipboard
--output, -o TEXT    Write output to file

# Search command options
--attendees, -a      Search participant names
--transcript, -t      Search transcript content (slower)
--show-ignored        Include ignored meetings
--include-empty-transcripts Include meetings with no transcript content (default: False)
--json               Output results in JSON format
--yaml               Output results in YAML format

# Clean command options
--dry-run            Show what would be removed (default)
--confirm            Actually mark meetings for removal
--min-words INTEGER  Minimum word count threshold (default: 10)
--min-duration INTEGER Minimum duration in minutes
--reset              Reset ignore list

Available MCP Tools

search_meetings

Search meetings by query string.

Parameters:
- query (string): Search query for meetings
- limit (integer, optional): Maximum number of results (default: 10)

get_meeting_details

Get detailed information about a specific meeting.

Parameters:
- meeting_id (string): Meeting ID to retrieve details for

get_meeting_transcript

Get transcript for a specific meeting.

Parameters:
- meeting_id (string): Meeting ID to get transcript for

get_meeting_documents

Get documents associated with a meeting.

Parameters:
- meeting_id (string): Meeting ID to get documents for

analyze_meeting_patterns

Analyze patterns across multiple meetings.

Parameters:
- pattern_type (string): Type of pattern to analyze ('topics', 'participants', 'frequency')
- date_range (object, optional): Date range for analysis with start_date and end_date

Usage Examples

Once configured with Claude Desktop, you can use natural language to interact with your Granola meetings:

Basic Queries

  • "Search for meetings about quarterly planning"
  • "Show me yesterday's meetings"
  • "Find meetings with David from this week"

Transcript Access

  • "Get the transcript from yesterday's ProofChat meeting"
  • "What was discussed in the Float rollback planning meeting?"
  • "Show me the full conversation from the David Tibbi meeting"

Content Analysis

  • "Analyze participant patterns from last month"
  • "What documents are associated with the product review meeting?"
  • "Search for mentions of 'schema labeling' in meeting transcripts"

Recent Meeting Intelligence

The server automatically detects and provides access to:

  • Full transcripts from recent meetings (25,000+ characters)
  • Meeting content including notes and summaries
  • Participant information and speaker identification
  • Local timezone display for all meeting times

Development

Running the CLI

# List meetings
uv run granola list

# Get meeting details
uv run granola get

# Clean up meetings
uv run granola clean

Running the MCP Server Directly

uv run granola-mcp-server

Adding Dependencies

uv add package-name

Configuration

Claude Desktop Config Locations

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json
  • Linux: ~/.config/Claude/claude_desktop_config.json

Granola Cache Location

The server reads from Granola's cache file at:

~/Library/Application Support/Granola/cache-v3.json

Security & Privacy

  • ✅ 100% Local Processing - All data stays on your machine
  • ✅ No External API Calls - No data sent to external services
  • ✅ Granola Permissions Respected - Uses existing Granola.ai access controls
  • ✅ Read-Only Access - Server only reads from Granola's cache

Performance & Capabilities

  • Fast Loading: Sub-2 second cache loading for hundreds of meetings
  • Rich Content: Extracts 25,000+ character transcripts and meeting notes
  • Efficient Search: Multi-field search across titles, content, participants, and transcripts
  • Memory Optimized: Lazy loading with intelligent content parsing
  • Timezone Smart: Automatic local timezone detection and display
  • Production Ready: Successfully processes real Granola data (11.7MB cache files)
  • Scalable: Handles large datasets with 500+ transcript segments per meeting

Current Status

🚀 PRODUCTION READY - Successfully tested with real Granola.ai data including:

  • ✅ 39+ meetings parsed and searchable
  • ✅ 28 full transcripts with complete conversations
  • ✅ Rich meeting content from notes, summaries, and structured data
  • ✅ Timezone intelligence showing times like "17:04" instead of "21:04 UTC"
  • ✅ Speaker identification and conversation flow
  • ✅ Yesterday's meetings fully accessible with detailed content

Troubleshooting

Common Issues

"Cache file not found"

  • Ensure Granola.ai is installed and has processed some meetings
  • Check that the cache file exists: ls -la "~/Library/Application Support/Granola/cache-v3.json"

"uv command not found"

  • Install uv: curl -LsSf https://astral.sh/uv/install.sh | sh
  • Or use pip fallback in Claude config: "command": "python"

"Permission denied"

  • Ensure the cache file is readable: chmod 644 "~/Library/Application Support/Granola/cache-v3.json"

Server not appearing in Claude Desktop

  • Verify the absolute path in your Claude config
  • Check Claude Desktop logs for MCP server errors
  • Restart Claude Desktop after config changes

CLI-Specific Issues

"No meetings found"

  • Run uv run granola list to see if meetings are loaded
  • Check cache file size: ls -lh "~/Library/Application Support/Granola/cache-v3.json"
  • Verify Granola.ai has processed recent meetings

"Interactive selection not working"

  • Ensure you're running in a terminal that supports interactive prompts
  • Try using a specific meeting ID: uv run granola get <meeting-id>

"Clipboard not working"

  • On macOS, ensure pbcopy is available: which pbcopy
  • Use file output as fallback: uv run granola get --output file.md

"Clean command shows no results"

  • Check current ignore list: uv run granola list --show-ignored
  • Reset ignore list: uv run granola clean --reset
  • Adjust thresholds: uv run granola clean --min-words 5