Fellow

Local MCP server wrapping Fellow.ai API with a cache, sync, and search for meetings data
  • javascript

0

GitHub Stars

javascript

Language

6 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "liba2k-fellow-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "fellow-mcp"
      ],
      "env": {
        "FELLOW_API_KEY": "YOUR_FELLOW_API_KEY_HERE",
        "FELLOW_SUBDOMAIN": "YOUR_SUBDOMAIN"
      }
    }
  }
}

You run a local MCP server that wraps the Fellow.ai API to access meetings, transcripts, summaries, action items, and participants. It caches data locally, keeps action items fresh, and lets you search cached notes efficiently from your MCP client.

How to use

Install, configure, and run the Fellow MCP server to enable your MCP client to query meetings, transcripts, summaries, action items, and participants. You can perform incremental syncing to keep data up to date and use full-text search to find information across cached notes. Use the MCP client to search meetings by participant, fetch transcripts, generate summaries, extract action items, and retrieve attendees.

How to install

Prerequisites: Node.js version 18 or newer, and an active Fellow.ai account with API access.

npm install -g fellow-mcp

Notes on configuration and usage

Set up your Fellows API credentials and point your MCP client to the local server. The server runs locally as a stdio-based process and reads API credentials from environment variables.

{
  "mcp": {
    "fellow": {
      "type": "stdio",
      "command": ["npx", "-y", "fellow-mcp"],
      "environment": {
        "FELLOW_API_KEY": "YOUR_FELLOW_API_KEY_HERE",
        "FELLOW_SUBDOMAIN": "YOUR_SUBDOMAIN"
      },
      "enabled": true
    }
  }
}

Operational notes

  • The local SQLite cache stores notes, recordings, action items, and participants for fast access and offline usage.

  • You can request incremental sync by default. Use a full re-sync if you need to refresh everything.

  • Use the API tools via your MCP client to search meetings, fetch transcripts, generate summaries, and query action items and participants.

Security and access

Keep your API key secure. Do not expose the FELLOW_API_KEY and FELLOW_SUBDOMAIN in client-side code or public repositories. Use environment variable management best practices for your deployment environment.

Available tools

search_meetings

Search for meetings or recordings in Fellow with optional filters for title, creation date range, and result limit.

get_meeting_transcript

Retrieve the full transcript of a meeting recording, including speaker labels and timestamps.

get_meeting_summary

Obtain the meeting summary including agenda items, discussion topics, and decisions.

get_action_items

Extract action items from a meeting's notes, with optional assignee and date filtering.

get_meeting_participants

Fetch the list of participants/attendees for a meeting.

sync_meetings

Synchronize meetings from Fellow API to the local SQLite cache, with incremental or full sync options.

get_all_action_items

Query all action items from the local cache, automatically performing an incremental sync first.

get_meetings_by_participants

Find meetings that included specified participants based on their email addresses.

search_cached_notes

Full-text search across all cached meeting notes and content.

get_sync_status

Retrieve the current sync status and database statistics.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational