Connect to Fellow to pull meeting recordings, notes, transcripts, and workspace details directly into your workflows. Search, filter, and retrieve the exact meetings or notes you need, then pass the structured outputs to follow-up steps like summaries, action items, or reporting.
Instantly access meeting recordings, notes, and transcripts across your Fellow workspace with structured outputs ready for automation.

​
What is Fellow MCP?

The Fellow MCP creates a customized node that understands Fellow’s API so you can interact with your meeting data using natural language. You write a clear prompt, the AI generates a reusable node tailored to Fellow, and it returns structured data that fits perfectly into automated workflows.

​
What Can It Do for You?

  • List meeting recordings with filters by date or keyword, and optionally include transcripts
  • Retrieve a specific recording by ID with transcript details for downstream analysis
  • List meeting notes with filters, and optionally include full content and attendees
  • Get authenticated user and workspace information to power routing and governance

​
Available Tools

ToolWhat It DoesExample Use
Get MeGet information about the authenticated user and their workspace”Get my account and workspace details and return structured data with user name, user email, workspace name, and workspace id”
List RecordingsList meeting recordings with filtering options and optional transcript inclusion”List recordings between start date and end date that match search term, include transcripts if available, and return structured data with recording id, title, meeting date, duration, and has_transcript”
Get RecordingRetrieve a specific recording by its ID with optional transcript data”Using recording id, get the recording, include transcript, and return structured data with title, meeting date, duration, and transcript text”
List NotesList meeting notes with filtering options, and optionally include content and attendees”List notes between start date and end date that match search term, include content and attendees, and return structured data with note id, meeting title, meeting date, attendees, and content”
Get NoteRetrieve a specific note by its ID with optional content and attendees”Using note id, get the note, include content and attendees, and return structured data with meeting title, meeting date, attendees, and full note content”

​
How to Use

1

Create Your Fellow MCP Node

Go to your node library, search for Fellow, and click “Create a node with AI”
2

Add Your Prompt

Drag the Fellow MCP node to your canvas and add your prompt in the text box.
3

Test Your Node

Run the node to see the results. If it works as expected, you’re all set! If you run into issues, check the troubleshooting tips below.
4

Save and Reuse

Once your Fellow MCP node is working, save it to your library. You can now use this customized node in any workflow.

​
Example Prompts

Here are some prompts that work well with Fellow MCP: Find Recordings: 
List recordings between `start date` and `end date` that match `search term`, do not include transcripts, and return structured data with recording id, title, meeting date, and duration
Retrieve a Recording With Transcript: 
Using `recording id`, get the recording, include transcript, and return structured data with title, meeting date, duration, and transcript text
Review Notes With Attendees: 
List notes between `start date` and `end date` that match `search term`, include content and attendees, and return structured data with note id, meeting title, meeting date, attendees, and content
Workspace Info For Governance: 
Get my account and workspace details and return structured data with user name, user email, workspace name, and workspace id
Start simple and focused. Use List Recordings or List Notes to get IDs first, then use Get Recording or Get Note to pull full details. Fellow MCP works best with focused, single-action prompts, and passing structured data between steps keeps workflows reliable.

​
Troubleshooting

If your Fellow MCP node isn’t working as expected, try these best practices:

​
Keep Prompts Simple and Specific

  • Good: “Search for meeting title keyword between start date and end date and return structured data with recording id and meeting date”
  • Bad: “Find all recordings this month, summarize each transcript, and get the notes”
While the bad prompt might eventually work, a more efficient approach is to break tasks into separate nodes and keep each Fellow prompt focused on data retrieval.

​
Match What Fellow Can Do

  • Good: “Using note id, get the note, include content and attendees, and return structured data with meeting title, meeting date, attendees, and content”
  • Bad: “Send a follow-up email to all attendees with a summary of the note content”
Fellow MCP excels at retrieving meeting data. For emailing, combine it with Gmail Sender or another messaging node.

​
Break Complex Tasks Into Steps

Instead of trying to do everything in one prompt (which might cause timeouts or errors): 
List all recordings in `date range`, fetch transcripts, summarize each one, and post summaries to `slack channel`
Break this into smaller, focused nodes:
1

Step 1: List Recordings

List recordings between start date and end date that match search term and return structured data with recording ids and titles
2

Step 2: Get Recordings + Transcripts

Using recording ids, get each recording with transcript and return structured data with title, meeting date, duration, and transcript text
3

Step 3: Summarize and Share

Use Ask AI to summarize transcript text with summary style, then send the summaries to slack channel using Slack Message Sender
Connect these nodes sequentially in your workflow. Outputs from each step feed directly into the next, creating a smooth, automated process.

​
Focus on Data Retrieval

Fellow MCP is ideal for accessing information from Fellow. For analysis or content creation, link it to other nodes. Example:
  • Good prompt: “List notes matching search term between start date and end date and return structured data with note id, meeting title, and meeting date”
  • Bad prompt: “List notes, extract action items, and draft follow-up messages”
For analysis and drafting content, use Ask AI or similar nodes. Fellow MCP works best when focused on retrieving structured data.

​
Troubleshooting Node Creation

If you’re seeing empty outputs in the node creation window (or if you’ve already created or saved the node, hover over it and click “Edit”), use the chat interface to prompt the AI to add debug logs and verify the API response. Specifically mention that you received empty outputs.
In the node creation window (or if you’ve already created or saved the node, hover over it and click “Edit”), use the chat interface to describe in detail what you expected versus what you received.
First click “Fix with Gummie”. If multiple attempts fail, simplify your prompt or contact support.
MCP node creation often requires a few tweaks. Use the chat interface in the node creation window to refine filters, output fields, or pagination. The AI will adjust the node based on your feedback.

​
Need More Help?