Read Excel Files With AI on Mac (excel_read)

The excel_read tool lets your AI assistant open and read any Excel .xlsx workbook stored on your Mac. It returns the sheet names and the actual cell values, so the AI can reason over your numbers without you ever copy-pasting a table. You just point it at a file and ask a plain-English question. For example, you can type: "Read ~/Downloads/Q2-budget.xlsx and tell me which department went over budget." The AI loads the workbook, scans the cells, and answers from your real data.

This is a read-only tool, so it never changes your spreadsheet. It simply surfaces what's already there: the rows, the columns, the totals, the formulas' results. Install LMCP to get started: Download LMCP.

Which AI agents work?

excel_read works with every major AI client through LMCP. Desktop clients auto-configure over local stdio, and 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 AIs (via the LMCP Cloud Relay connector): ChatGPT, Claude.ai (web), Grok, and Perplexity.

The same excel_read tool is exposed to all of them, alongside 150+ other LMCP tools for your Mac apps.

Automation

excel_read becomes powerful when it chains with other tools. Ask the AI to read a sales workbook and then draft an email summary with send_email, or read a budget sheet and create calendar reminders for upcoming payment dates with create_calendar_event. You can read figures from one .xlsx, have the AI compute new totals, and write them back into a fresh workbook with excel_write. Combine it with finder_search to locate every spreadsheet in a folder, then read and compare them in one pass.

Context

Because LMCP runs natively on your Mac, the AI works with your actual files, not a generic example. It reads the exact workbook sitting in your Downloads, Documents, or a synced OneDrive or Google Drive folder. There is no upload step and no stale copy. The AI sees the same numbers you see when you open the file in Excel or Numbers, so its answers reflect your real, current data.

Productivity

Reading spreadsheets manually is slow: scrolling across tabs, hunting for a cell, eyeballing trends. With excel_read, a question like "which line items grew more than 10% quarter over quarter?" gets answered in seconds. You skip the copy-paste into a chat window, skip reformatting, and skip the transcription errors. A monthly reporting task that took 20 minutes of clicking turns into one prompt. Multiply that across every workbook you touch and the saved time adds up fast.

Privacy & GDPR

Everything happens on your Mac. excel_read opens the file locally and passes only what's needed back to your AI client; the workbook is never uploaded to an LMCP server and nothing is stored in the cloud. There are no API keys and no third-party data processor in the loop for the file itself. This local-first design makes LMCP GDPR-compliant by architecture — your financial and personal data stays under your control.