← All integrations
Custom MCP
AI assistant integration
Live
Included on
Pro, Team & Enterprise

Custom MCP

Use TeamsMaestro in any AI tool that supports a remote MCP server, including Cursor, Perplexity, and VS Code. Add one server URL, sign in with your TeamsMaestro account, and ask about your meetings without leaving the tool you work in.

Your AI tool may also need a paid plan, or an admin who allows custom MCP servers.

How to connect

Every AI tool puts its MCP settings in a slightly different place. The setup itself is the same everywhere.

  1. Open your AI tool's settings and find the section called MCP, Connectors, or Tools.
  2. Add a custom or remote MCP server, name it TeamsMaestro, and paste this server URL.
  3. Sign in with your TeamsMaestro account when the tool asks, and approve access.
https://mcp-tm.maestrolabs.com/mcp

Then ask a question that needs your meetings: "What did we decide on yesterday's pricing call?" or "List my open action items from this week."

Cursor. Add the server from Cursor's MCP settings, or put it in ~/.cursor/mcp.json to use it in every project (.cursor/mcp.json for one project only).

{
  "mcpServers": {
    "teamsmaestro": {
      "url": "https://mcp-tm.maestrolabs.com/mcp"
    }
  }
}

Perplexity. Open Plugins in the sidebar, or Customize and then Connectors. Choose Add custom, then Connector, then Add MCP connector. Pick Remote, paste the server URL, set Authentication to OAuth 2.0, and select Add. On a company account, an admin may need to allow custom connectors first.

VS Code with GitHub Copilot. Run MCP: Add Server from the Command Palette and paste the server URL, or add it to .vscode/mcp.json in your workspace. VS Code opens the TeamsMaestro sign-in in your browser the first time it connects.

{
  "servers": {
    "teamsmaestro": {
      "type": "http",
      "url": "https://mcp-tm.maestrolabs.com/mcp"
    }
  }
}

Settings screens change often. If yours looks different, the TeamsMaestro MCP setup guide walks through the general steps. The server URL stays the same.

Technical reference

The server URL is https://mcp-tm.maestrolabs.com/mcp. It uses the Streamable HTTP transport.

Sign-in uses OAuth 2.0 with PKCE and dynamic client registration, so your AI tool registers itself the first time it connects. Leave any client ID and client secret fields empty.

Everything the server exposes today is read-only. Your AI tool can look at your meetings and answer questions about them. It cannot create, edit, or delete anything in TeamsMaestro.

list_recent_meetings

Lists meetings from the last N days.

Read
list_upcoming_meetings

Lists scheduled meetings in the days ahead.

Read
get_meeting_summary

Returns the AI summary for a single meeting.

Read
get_meeting_transcript

Returns the full transcript for a single meeting, with paging for long calls.

Read
get_meeting_action_items

Returns action items, either for one meeting or across recent meetings.

Read
search_meetings

Searches meetings by keyword, with an optional date range.

Read
list_tags

Lists the tag names you put on meetings, so searches can filter by tag.

Read
get_product_faq

Answers questions about TeamsMaestro itself.

Read

Frequently asked questions

Which AI tools work with the TeamsMaestro MCP server?

Any tool that can connect to a remote MCP server over HTTPS and sign in with OAuth. That includes Cursor, Perplexity, and VS Code with GitHub Copilot. Claude, ChatGPT, and Microsoft Copilot have their own pages with app-specific steps: Claude, ChatGPT, and Microsoft Copilot.

Do I need an API key?

No. You sign in with your TeamsMaestro account and approve access, and the tool handles the rest. If it asks for a client ID or client secret, leave both empty.

Do I need a paid TeamsMaestro plan?

Yes. Connecting over MCP is included on the Pro, Team, and Enterprise plans. It is not available on the Free plan. Some AI tools also limit custom MCP servers to their own paid plans.

Who can see my meeting data?

Only the AI tool you connect, and only when you ask it something that needs your meetings. Once your tool reads that data, its own data and training policy applies, so check its privacy settings. Maestro Labs' own LLM providers are contractually barred from training on your data.

My AI tool has no option for a custom MCP server. What now?

Some tools only offer custom connectors on certain plans, and on company accounts an admin may need to turn them on. If your tool does not support remote MCP servers yet, Zapier or webhooks can send your meeting notes to it instead.