Where do I get the contact ID that m365_get_contact needs?

Run m365_list_contacts or m365_search_contacts first. Each result includes the contact's ID, which your AI passes straight into m365_get_contact — usually all in one conversation without you copying anything.

Does this send my Microsoft 365 contacts to a server?

No. LMCP runs on your Mac and queries Microsoft 365 directly with your own authenticated connection. The contact details are returned to your AI client locally and never stored or copied to any LMCP cloud server.

Can web AI assistants like ChatGPT and Perplexity use this tool?

Yes. Desktop clients (Claude Desktop, Cursor, VS Code, Windsurf, Zed) connect automatically over local stdio, while web AIs (ChatGPT, Claude.ai, Grok, Perplexity) connect through the LMCP Cloud Relay connector.

Get a Microsoft 365 Contact by ID with AI on Mac

The m365_get_contact tool lets your AI assistant fetch the full details of a single Microsoft 365 contact by its unique ID — name, every email address, phone numbers, job title, company, and address fields. You already have the ID from a previous lookup, and this tool expands it into the complete record. A user simply types something like: "Get the full details for that Microsoft 365 contact I just found and draft an intro email to them."

Because it works on the contact ID, m365_get_contact is the precise, single-record companion to broader search tools — no guessing, no fuzzy matches, just the exact person you asked for. Download LMCP to enable it.

Which AI agents work?

LMCP exposes 150+ tools to any MCP-capable assistant. Desktop clients auto-configure over a local stdio connection: Claude Desktop, Cursor, VS Code (GitHub Copilot), Windsurf and Zed. Web-based AIs connect through the LMCP Cloud Relay connector: ChatGPT, Claude.ai (web), Grok and Perplexity.

Local stdio (auto): Claude Desktop, Cursor, VS Code, Windsurf, Zed
Cloud Relay connector: ChatGPT, Claude.ai, Grok, Perplexity

Automation

This tool shines as the second step in a chain. Your AI calls m365_list_contacts or m365_search_contacts to find the right person, grabs the contact ID, then hands it to m365_get_contact for the complete record. From there the AI can keep going: draft an email with send_email, schedule a meeting using the contact's address with a calendar tool, or cross-reference the company against your Mail history. The ID-based design makes these multi-step workflows reliable — the AI never loses track of which contact it's working with.

Context

Your AI is reading your real Microsoft 365 directory, live, through your own authenticated Graph connection on your Mac. There's no stale export, no copied spreadsheet, no manual paste. When a colleague updates their phone number or title in your org's directory, your assistant sees the current value the moment it asks. That real-time grounding is what turns a generic chatbot into an assistant that actually knows your contacts.

Productivity

Pulling one contact's full card normally means switching to Outlook on the web, searching, clicking into the person, and copying fields by hand. With m365_get_contact your AI does it inline in the conversation — and immediately uses the result. "Find Maria in our M365 directory, get her full details, and write a follow-up email to her work address" becomes a single request instead of five app switches. Over a busy day of outreach and scheduling, that saves real minutes per contact and keeps you in flow.

Privacy & GDPR

Everything runs locally on your Mac. LMCP talks to Microsoft 365 directly from your machine using your own credentials — your contact data is never copied to an LMCP server, never logged in the cloud, and never used for training. There are no third-party API keys to hand out and nothing stored remotely. This local-first design makes LMCP GDPR-compliant by architecture, because the personal data simply never leaves your control.