Does gdrive_file_info need a Google API key or OAuth?

No. It reads the Google Drive folder already synced to your Mac by the official Drive desktop app. There are no API keys, no OAuth, and no cloud calls — LMCP works entirely with the local files on disk.

How is this different from listing the whole folder?

Listing a directory scans every item, which is slow on large shared folders. gdrive_file_info targets one file or folder and returns just its size, dates, and type — much cheaper when you only need details about a single item.

Can web AIs like ChatGPT or Perplexity use this tool?

Yes. Desktop clients (Claude Desktop, Cursor, VS Code, Windsurf, Zed) auto-configure over local stdio, while web AIs (ChatGPT, Claude.ai, Grok, Perplexity) connect through the LMCP Cloud Relay connector to reach your Mac.

Get Google Drive File Info with AI on Mac

The gdrive_file_info tool returns the metadata for a single file or folder inside your synced Google Drive on the Mac: its size, creation and modification dates, and type. It reads the local Google Drive folder directly, so it's far cheaper and faster than listing an entire directory just to learn about one item. A natural prompt looks like: "How big is the Q3-budget.xlsx file in my Google Drive, and when was it last modified?" — and the AI answers with the exact byte size and timestamps, no browser tab or Drive web UI needed.

Because it's a read-only tool, it never changes anything. It's the metadata equivalent of a quick "get info" right-click — perfect when you only need to confirm a file exists, check if it's stale, or verify a download finished before acting on it.

Which AI agents work?

LMCP exposes gdrive_file_info to every MCP-capable assistant. Desktop clients auto-configure over local stdio the moment LMCP is installed:

Claude Desktop
Cursor
VS Code (GitHub Copilot)
Windsurf
Zed

Web-based AIs connect through the LMCP Cloud Relay connector, which securely bridges them to your Mac:

ChatGPT
Claude.ai (web)
Grok
Perplexity

Whichever you use, the file metadata is read locally on your machine and handed back to the model in plain language.

Automation

Metadata lookups are the glue between bigger workflows. Chain gdrive_file_info with related LMCP tools across 150+ available: list a folder to find a candidate file, call gdrive_file_info to confirm its size and modified date, then read or attach it to an email with the Mail tools. Or have the AI poll a file's modification date before triggering a downstream step — "if report.pdf was updated today, summarize it and send it to my team." The cheap metadata call keeps these multi-step routines fast and avoids re-scanning whole directories.

Context

The AI isn't guessing about some generic cloud file — it's working with your real, synced Google Drive on your actual Mac. That means correct sizes, accurate timestamps, and the true file type as it sits on disk right now. The assistant can reason about your genuine documents: "Is the contract folder bigger than usual?" or "Which of these two exports is newer?" answered from live local data, not a stale cache.

Productivity

Checking a file's size or last-modified date normally means opening Finder or the Drive web app, navigating folders, and right-clicking for Get Info. With gdrive_file_info you just ask. Over a busy day of verifying uploads, confirming the right version, or spotting stale files, that's dozens of context switches saved. The AI returns precise numbers in a single turn, and because it's cheaper than listing a directory, big shared folders respond instantly.

Privacy & GDPR

Everything runs locally on your Mac. gdrive_file_info reads the metadata from the Google Drive folder already synced to your disk — nothing is uploaded, copied to a server, or stored in the cloud by LMCP. There are no API keys and no third-party data processor. This is privacy by architecture: your files never leave your machine. Read more about GDPR-compliant by architecture.

Ready to try it? Download LMCP, a free native macOS MCP server with 150+ tools.