1. Conduid
  2. Science
  3. McpOffice
MCP server · Science

McpOffice

MCP server for Microsoft Office documents (C# + DevExpress.Docs)

Unclaimed science
34Low

Scored 5 months ago · breakdown

About McpOffice

McpOffice is an MCP server in the Science category: mCP server for Microsoft Office documents (C# + DevExpress.Docs).

Install

From source
git clone https://github.com/MBrekhof/mcpOffice

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

✦

Ask AI about McpOffice

Powered by Claude · Grounded in docs

I know everything about McpOffice. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

  • ✓README presentThe repository ships documentation.

5 checks have not been run yet.

README

mcpOffice

An MCP (Model Context Protocol) server for Microsoft Office documents, written in C# (.NET 9) and backed by DevExpress Office File API packages. It lets AI agents read, write, and convert Office documents through tool calls instead of one-off scripts.

Status: Word (.docx) and Excel (.xlsx / .xlsm) POCs are complete, including excel_analyze_vba v1 (procedures, event handlers, call graph, object-model references, file/DB/network/automation/shell dependencies). Next: excel_analyze_vba v2 — conversion hints toward Excel-to-C# migration tooling.

Architecture

mcpOffice architecture

Source: docs/img/architecture.excalidraw (open in Excalidraw). See ARCHITECTURE.md for the full layer map.

Documents

Current Tools

24 tools shipped: 1 ping + 15 Word + 8 Excel.

Word

  • word_get_outline(path)
  • word_get_metadata(path)
  • word_read_markdown(path)
  • word_read_structured(path)
  • word_list_comments(path)
  • word_list_revisions(path)
  • word_create_blank(path, overwrite=false)
  • word_create_from_markdown(path, markdown, overwrite=false)
  • word_append_markdown(path, markdown)
  • word_find_replace(path, find, replace, useRegex=false, matchCase=false)
  • word_insert_paragraph(path, atIndex, text, style?)
  • word_insert_table(path, atIndex, headers[], rows[][])
  • word_set_metadata(path, properties)
  • word_mail_merge(templatePath, outputPath, dataJson)
  • word_convert(inputPath, outputPath, format?)

Excel

  • excel_list_sheets(path) — sheets in order with visibility, used range, dimensions.
  • excel_read_sheet(path, sheetName?, sheetIndex?, range?, includeFormulas=true, includeFormats=false, maxCells=50000) — cell data with formulas + formats.
  • excel_get_metadata(path) — author, title, created/modified, sheet count, document properties.
  • excel_list_defined_names(path) — workbook + sheet-scoped names with refersTo / scope / hidden flag.
  • excel_list_formulas(path, sheetName?, includeValues=false, maxFormulas=10000) — formula cells with optional cached values.
  • excel_get_structure(path, includeSheets=true, includeFormulas=true, includeDefinedNames=true) — workbook rollup sized for huge workbooks.
  • excel_extract_vba(path) — static VBA module source via in-process MS-OVBA decompression (no Excel install required).
  • excel_analyze_vba(path, includeProcedures=true, includeCallGraph=false, includeReferences=false, moduleName?) — structural analysis on top of the extracted source: procedures with signatures, event handlers, call graph (with intra-workbook resolution), Excel object-model references, and external dependencies (file/DB/network/automation/shell). Pass moduleName to scope the heavy arrays to a single module on large workbooks; the summary stays whole-workbook.

Other

  • Ping — health check, returns pong.

All file paths passed to tools must be absolute.

Example

Create a Word document from Markdown, then convert it to PDF:

{
  "path": "C:\\Temp\\proposal.docx",
  "markdown": "# Proposal\n\nHello **Word**.",
  "overwrite": false
}
{
  "inputPath": "C:\\Temp\\proposal.docx",
  "outputPath": "C:\\Temp\\proposal.pdf"
}

Extract VBA modules from a macro-enabled workbook:

{
  "path": "C:\\Workbooks\\AnalysisTool.xlsm"
}

Roadmap

  1. Word POC — read / write / convert .docx ✓
  2. Excel POC — read sheets, list formulas/structure/defined names, extract VBA ✓
  3. excel_analyze_vba v1 — call graph, event handlers, Excel object-model refs, external dependencies ✓
  4. excel_analyze_vba v2 — conversion hints (procedure role classification, suggested C# equivalents, DOT/Mermaid call-graph rendering, cross-module coupling score).
  5. PowerPoint (.pptx).
  6. PDF.

Built With

  • ModelContextProtocol — C# MCP SDK.
  • DevExpress RichEdit / Spreadsheet / Office File API packages — server-side document APIs.
  • MarkdownToDocxGenerator — richer Markdown-to-DOCX import.
  • OpenMcdf — OLE compound file reader for VBA project extraction.

README copy captured when this server was indexed; it may lag the repository. The original is authoritative.

Questions

About McpOffice

How do I install McpOffice?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is McpOffice safe to use with an AI agent?

Its trust score is 34 out of 100 (low). It passes 1 of the 1 static security checks run so far. It has no ConduID identity yet, so agent calls to it are not receipted.

Is McpOffice still maintained?

Conduid hasn't recorded a commit date for this repository yet. Check the repository directly for recent activity.