Does teams_read_chat_messages need a Microsoft Graph API key?

No. LMCP reads Teams data directly from the app on your Mac, locally. There are no API keys, tokens, or admin setup required.

Can web AIs like ChatGPT and Claude.ai read my Teams chats?

Yes. Desktop clients connect via local stdio, while web AIs (ChatGPT, Claude.ai, Grok, Perplexity) connect through the LMCP Cloud Relay connector. Your Teams data is still read locally on your Mac.

Is my Teams conversation data sent to the cloud?

No. The tool reads messages locally on your Mac and the AI receives only what it needs to answer your prompt. Nothing is stored on an LMCP server or shared with third parties.

Read Microsoft Teams Chat Messages with AI on Mac

Read your Microsoft Teams chats with AI

The teams_read_chat_messages tool lets your AI assistant pull the actual messages from a Microsoft Teams chat or direct-message thread on your Mac. Instead of switching to Teams, scrolling, and re-reading a long back-and-forth, you just ask. For example, you can type: "Read my Teams DM with Sarah and tell me what she decided about the launch date." The AI fetches the real thread and answers in plain language.

LMCP is a free, native macOS MCP server that gives AI assistants access to your real Mac apps — including Microsoft Teams — running entirely on your machine. No API keys, no Graph tokens, no cloud copy. It reads Teams data locally, exactly as you see it. Download LMCP to get started.

Which AI agents work?

LMCP works with every major AI client. Desktop apps auto-configure through a local stdio connection, and web-based AIs connect securely through the LMCP Cloud Relay connector.

Desktop (local stdio, auto-configured): Claude Desktop, Cursor, VS Code (GitHub Copilot), Windsurf, and Zed.
Web (via LMCP Cloud Relay): Claude.ai, ChatGPT, Grok, and Perplexity.

Once LMCP is installed and your client is restarted, teams_read_chat_messages appears alongside 150+ other tools your AI can call on demand.

Automation

Reading a chat is most powerful when chained with other tools. Your AI can read a Teams thread, then act on it without you copy-pasting anything:

Read a project DM, then create a reminder for the action item buried in it.
Pull a decision from a chat, then draft an email summarizing it for stakeholders.
Read a thread, then add a calendar event for the meeting being scheduled.
Combine with teams_list_chats to find the right conversation first, then read it.

Context

Because LMCP reads your real Teams data, the AI answers from your genuine conversations — not guesses or generic templates. It knows who said what, when, and in which thread. That means follow-up questions like "who pushed back on the timeline?" or "what was the final number we agreed on?" get accurate, grounded answers drawn from the actual message history on your Mac.

Productivity

A single Teams thread can hold a dozen decisions scattered across days. Manually re-reading to catch up after a day off, or to write a status update, eats real time. With teams_read_chat_messages, your AI digests the whole thread in seconds and surfaces only what matters. Catching up on a chat, extracting action items, or summarizing a discussion for a teammate drops from ten minutes of scrolling to a one-line prompt.

Privacy & GDPR

Everything runs locally on your Mac. LMCP reads Teams messages directly from the app on your machine — nothing is uploaded, stored on a server, or sent to a third party. There are no API keys to manage and no cloud copy of your conversations. This local-first design makes LMCP GDPR-compliant by architecture: your data never leaves your device, so there is no transfer or external processing to worry about.