Does finder_search look at file contents or just names?

It matches files by name across your filesystem using Spotlight/mdfind. To work with what's inside a file, pair it with LMCP's read tools, which can open and summarize PDFs, Word docs and spreadsheets.

Is my data sent to the cloud?

No. finder_search runs Spotlight locally on your Mac. File names and paths stay on your device, nothing is stored on a server, and no API keys are required.

Which AI assistants can use it?

Claude Desktop, Cursor, VS Code (GitHub Copilot), Windsurf and Zed connect locally via stdio. ChatGPT, Claude.ai, Grok and Perplexity connect through the LMCP Cloud Relay connector.

Find Any File on Your Mac with AI: finder

The finder_search tool lets your AI assistant find files anywhere on your Mac by name. It uses macOS Spotlight (the same mdfind index behind Cmd+Space), so results come back instantly without crawling your disk. You just describe what you're looking for in plain language and the AI runs the search for you.

For example, you can type: "Find the invoice PDF I saved last month called 'Acme'" and the assistant returns the matching file paths, ready to open, read or act on. No more digging through nested folders or trying to remember where you put something.

Because finder_search matches by file name across your whole filesystem, it's great for those moments when you know the document exists but have no idea which folder it lives in. Download LMCP to enable it.

Which AI agents work?

LMCP ships 150+ tools and works with every major AI assistant. Desktop clients auto-configure through a local stdio connection; web AIs 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.

Whichever you use, finder_search runs the same Spotlight query on your real Mac.

Automation

File search is most powerful when it feeds the next step. finder_search chains naturally with the rest of LMCP's 150+ tools:

Find a file, then have the AI read it (PDF, Word, Excel) and summarize the contents.
Locate a report and attach it to a new email or Message draft.
Search for a contract, then move or rename it through Finder tools.
Find a spreadsheet and pull a value into a calendar event or reminder.

The AI decides the sequence, so a single request like "find my Q3 budget and email it to my accountant" becomes one smooth flow.

Context

Unlike a generic chatbot, the AI here works with your actual Mac data. finder_search queries the live Spotlight index on your machine, so it sees the real files you have right now — the latest downloads, the project folder you created this morning, files on external drives that Spotlight has indexed. The assistant answers about your filesystem, not a guess.

Productivity

Manually hunting for a misplaced file can eat several minutes and break your focus. With finder_search, you stay in your AI conversation and get the path in seconds. Combine that with read-and-summarize and you go from "where is that file?" to "here's what it says" in a single prompt — saving the constant context-switching between Finder windows, Spotlight and your editor.

Privacy & GDPR

Everything runs locally on your Mac. finder_search calls Spotlight on your own device; file names and paths never leave your machine, and nothing is stored on any server. There are no API keys and no cloud copy of your data. This local-first design makes LMCP GDPR-compliant by architecture. Download LMCP and keep your files private.