How do I find the right OneDrive path to list?

Use the onedrive_root tool first — it returns your valid OneDrive roots (such as Personal and Business). Then point onedrive_list_files at any subfolder under one of those roots, like Documents/Invoices.

Does onedrive_list_files upload my files anywhere?

No. It only reads the folder listing locally on your Mac and returns the names and folders to your AI client. LMCP stores nothing on a server and uses no cloud copy of your data.

How many entries does it return?

Up to 1000 entries by default (a configurable limit). For very large folders you can raise or lower the limit, and the AI can drill into subfolders to narrow results.

List OneDrive Files With AI on Mac

The onedrive_list_files tool lets an AI assistant browse a folder in your OneDrive directly on your Mac. It returns the files and folders in any path you point it at — up to 1000 entries by default — so the AI can see what you actually have without you opening Finder or the OneDrive web app. A typical prompt is: "List the files in my OneDrive Documents/Invoices folder and tell me which ones were modified this month." The AI calls onedrive_list_files, reads the folder contents, and answers in plain language.

Not sure of the exact path? Pair it with onedrive_root to discover your valid OneDrive roots (Personal, Business, etc.), then drill into any subfolder. Download LMCP to add this and 150+ tools to your AI in a couple of minutes.

Which AI agents work?

LMCP works with every major MCP-capable assistant. Desktop clients auto-configure over a local stdio connection; web assistants connect 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 connector): ChatGPT, Claude.ai, Grok and Perplexity.

Once LMCP is installed, the same onedrive_list_files tool is available across all of them.

Automation

Listing files is the starting point for richer workflows. The AI chains onedrive_list_files with other tools: use onedrive_root to find the path, list the folder, then read or move specific files, attach one to an email with the Mail tools, summarize a document, or cross-reference filenames against a Calendar event or a Reminders list. Because the AI can enumerate folder contents first, it picks the right file instead of guessing — turning a vague request like "grab the latest contract" into a precise action.

Context

The AI works with your real OneDrive data on your Mac — the same files you see in Finder, including your synced Personal and Business accounts. Nothing is mocked or approximated. When you ask about "the proposal I saved last week," the assistant lists the actual folder, sees the real filenames and timestamps, and reasons over what is genuinely there. That grounding is what makes the answers trustworthy instead of generic.

Productivity

Browsing a deeply nested OneDrive folder by hand — opening Finder, expanding directories, scanning dozens of filenames — eats minutes every time. With onedrive_list_files you just ask: "What's in my OneDrive/Projects/2026 folder, sorted by what changed recently?" and get an instant, readable answer. Multiply that across daily file hunts, inventory checks, and cleanup tasks and it saves real time, while keeping you in your AI chat instead of switching apps.

Privacy & GDPR

Everything runs locally on your Mac. LMCP is a native macOS MCP server — it reads your OneDrive folder on the device and returns results to your AI client. No API keys, no copies uploaded to a server, nothing stored in the cloud by LMCP. Your file listings never leave your machine except as part of your own AI conversation. This is privacy by design — see GDPR-compliant by architecture for the full picture.