The MeetGeek MCP Server (Self-Hosted) is an open-source, local MCP implementation that allows MCP-compatible AI tools to access your MeetGeek meeting data directly from your machine.
Unlike the cloud MCP, this version:
Runs entirely locally
Uses your MeetGeek API key
Requires Node.js and manual setup
Best suited for developers and advanced users.
👉 GitHub repository: https://github.com/meetgeekai/meetgeek-mcp-server
1. Why Use the Self-Hosted MCP?
Runs locally — no third-party services
Full control over execution
Configure once, use across multiple MCP clients
Ideal for custom or offline workflows
2. Requirements
Node.js v14 or newer
A valid MeetGeek API key (found under your MeetGeek Settings → Integrations → Public API)
(Optional) MEETGEEK_BASE_URL for custom deployments
3. Setup Steps
Clone the repository:
git clone https://github.com/meetgeekai/meetgeek-mcp-server
cd meetgeek-mcp-serverInstall dependencies and build:
npm install
npm run buildRun the server:
MEETGEEK_API_KEY=your_key_here node dist/index.js
4. Configure Your MCP Client
Add the MCP server to your client configuration. For example:
{
"mcpServers": {
"meetgeek": {
"command": "node",
"args": ["path/to/dist/index.js"],
"env": {
"MEETGEEK_API_KEY": "your_key_here",
"MEETGEEK_BASE_URL": "https://api.meetgeek.ai"
}
}
}
}
Different clients may require different formatting—check their docs. Once configured, restart the client to enable tools.
5. Tools You Can Use
Tool | Description |
| Retrieve a paginated list of the authenticated user’s past meetings, including IDs, titles, timestamps, and participant counts. |
| Retrieve a paginated list of meetings for a specific team, including meeting metadata and participant information. |
| Fetch full details for a specific meeting, including title, timestamps, duration, participants, recording status, and processing state. |
| Retrieve the full meeting transcript with speaker names, timestamps, and spoken text. Supports pagination for long meetings. |
| Get an AI-generated summary highlighting main topics, key decisions, action items, and next steps. |
| Retrieve AI-generated highlights such as key moments, decisions, and notable quotes with timestamps and speaker context. |
| Access AI-powered analytics including sentiment analysis, topic breakdowns, speaking-time distribution, and engagement metrics. |
| Upload an audio or video recording via a direct download link for asynchronous AI analysis, generating transcripts, summaries, highlights, and insights. |
6. Basic Troubleshooting
“Module not found” errors → run
npm installin the project root.
Invalid API key errors or exits → re-check your key under MeetGeek account settings.
Client doesn't see tools → ensure the server is running after build, and STDIO isn't redirected or suppressed.
7. Want More Info?
The repository README includes advanced instructions like:
Debugging and logging flags
Full TypeScript source code and example usage
You’ll find the complete guide and setup details in the GitHub repository: https://github.com/meetgeekai/meetgeek-mcp-server
TL;DR (Step by Step)
Clone the repo
npm install → npm run build
Launch server with your
MEETGEEK_API_KEY
Add entry to your MCP‑client config
Restart your AI tool → Ask questions directly in chat
Prefer a plug-and-play setup?
Check out our Public MeetGeek MCP guide to connect MeetGeek to tools like ChatGPT, Claude, and MCP-compatible IDEs via OAuth.
Explore our Support Center for further guides, frequently asked questions, and tips. Or use the in-app live chat to reach out to our support team 💬
