Fathom AI

However, I plan on adding very customizable tools to this mcp so it will be kept alive and expanded on once I have time or if anyone else contributes.

Connect Claude to your Fathom meetings, transcripts, and AI summaries.

That's it. Ask Claude about your meetings.

Organizations: Admins must add the connector via organization admin settings, not personal settings.

Perplexity Desktop and other SSE clients

For clients that use the legacy SSE transport (such as Perplexity Desktop for Mac), use the SSE endpoint:

Refer to your client's documentation for how to add a remote SSE connector.

Search Fathom meetings by title, meeting title, host name, host email, or attendee name/email. Automatically scans up to 5 pages of results. This is an MCP-native tool that performs client-side filtering since Fathom's API doesn't provide a search endpoint.

"Show me my meetings from last week"

"Get the transcript from my standup yesterday"

"Summarize my meeting with the design team"

Your Fathom data is never stored or exposed by this server.

- Pass-through architecture: This server acts as a secure proxy, your Fathom data flows directly from Fathom to Claude without being stored, cached, or logged
- Encryption at rest: The only stored data (OAuth tokens) is encrypted using AES-256-GCM before being written to the database
- HTTPS only: All communication between Claude, this server, and Fathom is encrypted in transit

This MCP server defines a custom scope calledfathom:readfor tokens it issues to Claude. This is not a Fathom API scope - it's specific to this MCP server to describe read-only access to your Fathom data.

The Fathom API itself only provides read access via itspublic_apiscope. Write operations (creating/editing meetings, transcripts, etc.) are not available in the Fathom API.

- search_meetingsperforms client-side filtering since Fathom's API doesn't provide a search endpoint. For users with many meetings, uselist_meetingswith date filters instead.
- You can always ask the LLM what query params are avaialable.

Fathom OAuth apps require HTTPS redirect URIs, so local development withhttp://localhostisn't possible. Deploy to a hosting provider to test.

Railway (recommended), Render, or any platform that provides:

3. Configure Environment Variables (locally and in Railway)

Set these in your hosting provider's dashboard (as well as your local .env file to test build and start commands locally before pushing changes)

Add your deployed URL as a custom connector in Claude Desktop:

For SSE clients (Perplexity Desktop, etc.):

npm run dev # Start dev server with hot reload only for testing npm run build # Build for production npm run start # Run production build npm run lint # Check for linting errors npm run lint:fix # Fix linting errors npm run typecheck # Run TypeScript type checking npm run ci # Run all CI checks (lint, typecheck, test, build) npm run format # Format code with Prettier npm run db:studio # Open Drizzle Studio for database inspection npm run db:generate # Generate migrations from schema changes npm run db:migrate # Run pending migrations npm run db:push # Push schema directly (dev only)

Tip: Runnpm run cibefore pushing to ensure your changes pass GitHub Actions.

For pre-release features, use the staging URLs:

https://fathom-mcp-staging.up.railway.app/mcp

https://fathom-mcp-staging.up.railway.app/sse

- Transcript vectorization— Enable vectorization of large transcripts so LLMs can parse and understand them more efficiently. Would be implemented as a stateless worker to ensure no user data is persisted.
- Action item aggregation— Aggregate action items across meetings with filters. "Show all my incomplete action items from this week."
- Meeting analytics— Calculate stats like total meeting time, meeting frequency, and top attendees.
- Speaker time analysis— Analyze transcripts to show who spoke most in a meeting.
- Meeting comparison— Compare two meeting summaries to highlight what changed over time.
- Fathom API changelog monitoring— Automated detection of Fathom API changes via GitHub Action that periodically checks their API reference and creates an issue if changes are detected.

Monitor team productivity, usage, and spending for AI assistants using the Cursor Admin API.

Connect your Limitless Pendant data to Claude and other LLMs using the Limitless API.

Predict and track AI ROI using Monte Carlo simulations, real-time industry benchmarks, and ML-powered insights.

Connect Claude, ChatGPT, and other AI assistants to your Speak AI workspace. Transcribe meetings, analyze media, extract insights, all through natural conversation.

Stubble learns how you work and organizes your day automatically. Export timesheets, surface insights, and supercharge your AI tools.

AI-powered wellbeing insights with real-time benchmarks and alerts.

Run structured multi-agent decision reviews with Codex, Claude Code, and adapter-based CLI agents.

Interact with task, doc, and project data in Dart, an AI-native project management tool