Does todo_list_tasks need Microsoft To Do API access or a login?

No. Microsoft To Do syncs into macOS Reminders, and LMCP reads those lists locally through native Reminders access. There are no API keys, Graph tokens, or Microsoft logins required.

Can it read regular Apple Reminders lists too?

Yes. The tool lists tasks from any list in macOS Reminders, whether it originated in Microsoft To Do or was created directly in the Reminders app.

Will it modify or complete my tasks?

No. todo_list_tasks is read-only — it only reports your tasks, due dates, and notes. Use LMCP's separate write tools if you want the AI to create or update reminders.

List Microsoft To Do Tasks with AI on Mac

The todo_list_tasks tool lets your AI assistant read every task from a Microsoft To Do list (or any list in macOS Reminders) directly on your Mac. Because Microsoft To Do syncs into Apple Reminders, LMCP can surface those tasks locally without touching the Microsoft Graph API. You just ask in plain language — for example: "Show me everything on my Work list in Microsoft To Do that's still open." The AI calls todo_list_tasks, reads the live list, and answers with your real tasks, due dates, and notes.

This is a read-only tool, so it never changes or deletes anything — it only reports what's there. That makes it safe to lean on dozens of times a day for status checks, planning, and triage.

Which AI agents work?

LMCP exposes todo_list_tasks to every major AI client:

Desktop apps (auto-configure via local stdio): Claude Desktop, Cursor, VS Code (GitHub Copilot), Windsurf, and Zed. They detect LMCP automatically and start using your tasks right away.
Web AIs (connect through the LMCP Cloud Relay connector): ChatGPT, Claude.ai, Grok, and Perplexity. Add the LMCP connector once and the same local tools become available in the browser.

Either way, the data stays on your Mac — the relay only forwards the request to your local server.

Automation

Reading your tasks is most powerful when it chains with other tools. todo_list_tasks pairs naturally with the rest of LMCP's 150+ tools:

List open tasks, then ask the AI to create calendar events for the ones with hard deadlines.
Pull your task list and have the AI draft and send emails chasing the people each task depends on.
Combine your To Do list with unread Mail and today's Calendar into a single morning brief.
Turn loose tasks into structured reminders with due dates and alarms so nothing slips.

Context

Generic chatbots guess. With LMCP, the AI works from your actual Mac data — the same Microsoft To Do and Reminders lists you see in the app, synced live through macOS Reminders. There's no copy-paste, no exporting, no stale snapshot. Ask "what's overdue?" and the answer reflects the list as it stands this second, including tasks added from your phone moments ago.

Productivity

Instead of opening the To Do app, scanning multiple lists, and mentally filtering, you ask one question and get a clean, prioritized answer. A typical "what should I focus on today?" review that took a few minutes of clicking becomes a two-second prompt. Multiply that across daily standups, end-of-day wrap-ups, and weekly planning and you reclaim real time — while keeping the AI grounded in tasks you trust.

Privacy & GDPR

Everything runs locally. todo_list_tasks reads your lists on your Mac through native macOS Reminders access — no API keys, no cloud account, and no copy of your tasks stored on any server. Web AIs reach the tool through a thin relay that forwards the request to your machine and returns the result; the data itself never lands in a database. This is privacy by architecture, not by policy. Read more about how LMCP is GDPR-compliant by architecture.

Download LMCP and connect your AI to your real task lists in minutes.