mcpbeat

Granola

glebis/granola

This skill should be used when importing, listing, or exporting Granola meeting recordings and transcripts. Queries Granola's Personal API to list meetings, extract transcripts, and export to Obsidian notes in Fathom-compatible format.

7k tokens
context cost
the whole folder, loaded on every use
16
files
ships runnable scripts
0
copies elsewhere
how many repositories repackaged it
337
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository
npx skills add https://github.com/glebis/claude-skills --skill granola

What comes with it

23 426 bytes besides the instruction
.claude-plugin/plugin.json
references/cache-structure.md
scripts/granola.py
scripts/sync.sh
scripts/tests/__init__.py
scripts/tests/fixtures/cache_object_state.json
scripts/tests/fixtures/cache_string_state.json
scripts/tests/test_cmd_export.py
scripts/tests/test_cmd_list.py
scripts/tests/test_cmd_show.py
scripts/tests/test_compute_duration.py
scripts/tests/test_extract_calendar_times.py
scripts/tests/test_extract_people.py
scripts/tests/test_format_time.py
scripts/tests/test_load_cache.py

The instruction itself

13 sections, as written by the author

Granola Meeting Importer

Query Granola via Personal API to list notes, view transcripts, and export to Obsidian vault in the same format as the Fathom skill.

Prerequisites

  • Granola Business or Enterprise plan (Personal API required)
  • API key in sops-encrypted ~/Brains/brain/.env.granola as GRANOLA_API_KEY=grn_...
  • No additional dependencies (uses stdlib only)

Usage

python3 ~/.claude/skills/granola/scripts/granola.py <command> [options]

Commands

| Command | Description |

|---------|-------------|

| list | List notes from Personal API |

| show <note_id> | Show note details (summary, attendees, optionally transcript) |

| export <note_id> | Export note to Obsidian markdown (Fathom-compatible format) |

Options

| Option | Applies to | Description |

|--------|-----------|-------------|

| --format text\|json | list, show | Output format (default: text) |

| --after <ISO date> | list | Filter notes created after date |

| --all | list | Paginate through all results |

| --transcript | show | Include transcript in output |

| --vault <path> | export | Obsidian vault path (default: ~/Brains/brain) |

| --output <path> | export | Custom output file path |

Examples

List recent meetings

python3 ~/.claude/skills/granola/scripts/granola.py list
python3 ~/.claude/skills/granola/scripts/granola.py list --format json
python3 ~/.claude/skills/granola/scripts/granola.py list --after 2026-05-01

Show note with transcript

python3 ~/.claude/skills/granola/scripts/granola.py show not_5FkswTp4Omkpm5
python3 ~/.claude/skills/granola/scripts/granola.py show not_5FkswTp4Omkpm5 --transcript --format json

Export to Obsidian

python3 ~/.claude/skills/granola/scripts/granola.py export not_5FkswTp4Omkpm5
python3 ~/.claude/skills/granola/scripts/granola.py export not_5FkswTp4Omkpm5 --vault ~/Brains/brain

Output Format

Exported notes match Fathom skill format for consistency:

---
granola_id: not_xxxx
title: "Meeting Title"
date: YYYY-MM-DD
participants: ['Name 1', 'Name 2']
duration: HH:MM
source: granola
---

# Meeting Title

## Summary
{AI-generated summary}

## Transcript
**Speaker Name**: What they said...

Files saved as: Sessions/YYYYMMDD-meeting-title-slug.md

API Details

  • Base URL: https://public-api.granola.ai/v1
  • Auth: Bearer token (Personal API key, never expires)
  • Rate limits: 25 req burst / 5 req/sec sustained
  • Important: API only returns notes with generated summaries — in-progress meetings won't appear

Known Limitations

  • No live/in-progress access — notes appear only after Granola generates the AI summary
  • No per-utterance speaker names — Granola provides source (microphone vs speaker) and optional diarization_label. Export assigns meeting owner to microphone utterances
  • Note IDs required — use list first to get not_xxxx IDs, then show/export

Integration

  • transcript-analyzer: After export, run transcript-analyzer on the output file for deeper analysis
  • Fathom skill: Granola exports use the same frontmatter and transcript format as Fathom exports, so downstream tools work with both

How to use it

Copy the folder

Take glebis/granola from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.