productivityoauth2

Excel

Microsoft spreadsheets.

Verdict

The Excel MCP connects your team's spreadsheets to Switchy's AI workspace. @mention it to read cell ranges, update values, create pivot tables, or run formulas without opening the file. Analysts pull live data into reports, project managers sync task lists, and finance teams automate recurring calculations. Because it uses OAuth2, you grant access once and the whole team shares the connection. The catch: complex macros and VBA scripts won't run through the MCP — you're working with raw data and formulas only.

Common use cases

  • Pull quarterly revenue into standup summaries
  • Update project timelines from task tracker
  • Generate pivot tables for board decks
  • Compare budget vs actuals in real time
  • Automate weekly sales report formatting

Integration

Vendor
Excel
Category
productivity
Auth
OAUTH2
Composio slug
microsoft_excel

Tools

Per-tool listings haven't synced yet for Excel. The connection itself works - your Space can already @-mention it. Tool descriptions will fill in on the next Composio ingest.

Setup

Setup guide

  1. 11. Open your Switchy workspace and navigate to Settings > Integrations. 2. Find Excel in the productivity category and click Connect. 3. You'll redirect to Microsoft's login page — sign in with the account that owns your workbooks. 4. Grant permissions for read and write access to your Excel files (Switchy requests Files.ReadWrite.All scope). 5. After authorizing, you'll return to Switchy and see Excel marked as connected. 6. Open any Space and type '@Excel show me the data in Sheet1 cells A1 through C10 from Q4-Budget.xlsx' to test. 7. If the MCP returns cell values, your connection works. 8. Invite teammates to the Space — they can now @mention Excel without setting up their own auth.

What teammates see: by default, memories from Excel are scoped to the Space (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.

Works well with

Top models

Compatibility data appears once enough Spaces have used this MCP together with a given model.

How Switchy teams use it

Not enough Spaces yet to publish anonymised usage stats (we require ≥ 50 Spaces per week).

Starter prompts

Read Budget Snapshot

@Excel read cells B2 through E15 from the 'Monthly Budget' sheet in our Q1-2024.xlsx file and show me total spend by category.
Open in a Space →

Update Task Status

@Excel in ProjectTracker.xlsx, find rows where column C says 'In Progress' and change them to 'Done' if column D shows a date before today.
Open in a Space →

Create Sales Pivot

@Excel open SalesData.xlsx and create a pivot table on a new sheet that sums revenue by Region in rows and Product in columns.
Open in a Space →

Compare Two Sheets

@Excel compare cells A1:D20 in the Forecast sheet against the same range in Actuals sheet of Finance-Q2.xlsx and list any values that differ by more than 10%.
Open in a Space →

Calculate Running Total

@Excel in Expenses.xlsx, insert a new column F with a running total formula that sums column E from row 2 down to the current row.
Open in a Space →

Example outputs

Illustrative - representative of the model's voice and quality, not literal recordings.

Prompt

@Excel show me the current values in cells A1 through D10 of the Q1_Budget worksheet

Output

The Q1_Budget worksheet contains a table spanning A1:D10. Column headers are Department, January, February, March. Rows include Engineering ($45K, $48K, $52K), Marketing ($32K, $35K, $38K), and Sales ($28K, $30K, $33K). Total row shows $105K, $113K, $123K across the three months. This is an illustrative example — actual cell values depend on your workbook state.

Notes

This read operation requires OAuth consent with Files.ReadWrite.All or similar scope. The MCP returns cell ranges as structured data the AI can parse. If your workbook has formulas, you'll see computed results, not the formula text unless explicitly requested.

Prompt

@Excel update cell B5 in the Q1_Budget sheet to 50000 and add a comment saying 'Revised after headcount change'

Output

Cell B5 in Q1_Budget has been updated to 50000. A threaded comment reading 'Revised after headcount change' has been attached to that cell. The workbook's last-modified timestamp reflects this change. This is an illustrative example — the actual operation writes directly to your Excel file in OneDrive or SharePoint.

Notes

Write operations are immediate and irreversible through the MCP. If multiple teammates have the file open, Excel's co-authoring will sync the change, but the MCP doesn't queue or stage edits. Always verify the target cell reference before confirming destructive updates.

Prompt

@Excel read the Sales_Data sheet and calculate which product category had the highest month-over-month growth rate in Q1

Output

After reading Sales_Data (columns: Category, Jan_Sales, Feb_Sales, Mar_Sales), the AI calculates growth rates. Electronics grew 18% Jan-Feb and 22% Feb-Mar (compound 43.96%). Home Goods grew 12% and 15% (compound 28.8%). Apparel grew 8% and 10% (compound 18.8%). Electronics had the highest month-over-month growth in Q1. This is an illustrative example — the AI performs the calculation after the MCP retrieves the raw data.

Notes

This synthesis workflow pairs the MCP's data-fetch capability with the AI's math reasoning. The MCP doesn't calculate growth rates itself — it returns cell values, and the AI does the analysis. Useful when you need insight beyond what Excel formulas provide, but requires the AI to hold the full dataset in context.

Use-case deep-dives

Budget reconciliation for finance teams

When Excel MCP makes sense for monthly close workflows

A 3-person finance team closing monthly books needs to pull actuals from their accounting system, compare them to budget spreadsheets in Excel, and flag variances over 10%. The Excel MCP is the right call here if those budget models live in shared workbooks with formulas the team updates each quarter. OAuth2 means each person authenticates once and the workspace can read/write cells without manual exports. The trade-off: if your budget logic is simple enough to rebuild in a database or if you're already using Google Sheets, the Excel MCP adds a dependency you don't need. But if Excel is the source of truth and the models are complex, this MCP saves 2-3 hours of CSV wrangling per close cycle. Worth it for teams running 12+ closes a year.

Sales pipeline reporting for RevOps

Excel MCP for deal-stage snapshots at small scale

A RevOps analyst at a 20-person sales org maintains a weekly pipeline report in Excel that stitches together CRM exports, quota attainment, and forecast adjustments. The Excel MCP works if the report template has stable cell references and the analyst wants to automate the refresh without rewriting the whole thing in SQL. OAuth2 handles the auth once, and the workspace can update the sheet on a schedule or on-demand during pipeline reviews. The boundary: if the report pulls from more than 3 data sources or if the team grows past 50 reps, Excel becomes the bottleneck and you should move to a BI tool. Below that threshold, the MCP keeps the workflow intact and cuts manual update time from 45 minutes to under 5.

Inventory tracking for small retail ops

When Excel MCP fits (and when it doesn't) for stock counts

A 2-location retail business tracks daily inventory in an Excel workbook with one sheet per SKU category. The store manager updates counts each morning and the owner reviews stock levels before placing orders. The Excel MCP is borderline here: it works if the workbook structure is simple (flat tables, no VBA macros) and the team wants to query current stock levels from Switchy without opening the file. OAuth2 means the workspace can read the latest counts and flag low-stock items in a standup thread. The deal-breaker: if the workbook has complex macros, pivot tables that refresh on open, or if the team needs real-time updates across locations, Excel isn't the right system and the MCP won't fix that. For static daily snapshots under 500 SKUs, it's a clean fit.

Frequently asked

What does the Excel MCP let me do in Switchy?

The Excel MCP connects your Microsoft Excel workbooks to Switchy's AI workspace. Your team can query spreadsheet data, update cells, create charts, and run calculations without leaving the conversation. It works with both Excel Online and files stored in OneDrive or SharePoint, so you're not switching between tabs to pull numbers or update forecasts.

Do I need a Microsoft 365 admin account to connect Excel?

No. Any team member with a Microsoft account and access to the workbooks you want to use can connect Excel via OAuth2. Switchy requests only the permissions needed to read and write your spreadsheets — you're not granting org-wide admin access. If your company restricts third-party app connections, check with IT before connecting.

Can the Excel MCP write formulas or just read cell values?

It can both read and write. You can ask Switchy to insert formulas, update ranges, add new sheets, or pull specific cell values into a conversation. The MCP doesn't execute VBA macros or interact with pivot table slicers directly, but it handles the core spreadsheet operations most teams use daily.

Why use this instead of just opening Excel in a browser tab?

You use this when the spreadsheet is part of a larger workflow. If you're discussing a budget in Switchy and need to update three cells based on the conversation, the MCP does it inline. You're not context-switching, copying numbers, or explaining to the AI what's in column F — it already knows.

Who on my team should connect the Excel MCP?

Whoever owns the workbooks your team references most. If your finance lead maintains the budget sheet, they should connect it. If multiple people edit the same file, any of them can connect — Switchy respects Excel's existing share permissions, so you won't accidentally expose data someone shouldn't see.

Data last verified 7 hours ago.Sources aggregated hourly to weekly. See docs/architecture/model-directory.md.