Local MCP for Word, Excel, PowerPoint & PDF on macOS — formulas, slides, DOCX, PDF ops. No network.
MCPpedia last refreshed this data
io.github.JoyTruepath/truepath-office-mcp is an MCP server that local MCP for Word, Excel, PowerPoint & PDF on macOS — formulas, slides, DOCX, PDF ops. No network. Its tool list has not been published yet over stdio, requires no API key, and scores 40/100 on MCPpedia's security, maintenance and efficiency rubric.
Config is the same across clients — only the file and path differ.
{
"mcpServers": {
"truepath-office": {
"command": "/path/to/truepath-office-mcp/.build/release/truepath-office-mcp"
}
}
}Are you the author?
Add this badge to your README to show your security score and help users find safe servers.
Let your AI read and edit Office documents locally. Files never leave your Mac.
No automated test available for this server. Check the GitHub README for setup instructions.
Five weighted categories — click any category to see the underlying evidence.
No known CVEs.
No package registry to scan.
Be the first to review
Have you used this server?
Share your experience — it helps other developers decide.
Sign in to write a review.
Others in writing
MCP server for document format conversion using pandoc.
This is an MCP server that allows you to directly download transcripts of YouTube videos.
Scrape, crawl, and map websites to Markdown or JSON via local CLI.
Any URL to clean, LLM-ready Markdown for RAG. Strips ads, nav, and boilerplate.
MCP Security Weekly
Get CVE alerts and security updates for io.github.JoyTruepath/truepath-office-mcp and similar servers.
Start a conversation
Ask a question, share a tip, or report an issue.
Sign in to join the discussion.
truepath-office-mcp)Let your AI read and edit Office documents locally. Files never leave your Mac.
An MCP server that lets your AI agent — Claude Desktop, Claude Code, Codex, Cursor, Windsurf, or any MCP client — process Word (DOCX), Excel (XLSX), PowerPoint (PPTX) and PDF files on your machine: evaluate spreadsheet formulas, read & write workbooks and slide decks, extract structured DOCX content, and merge / split / protect PDFs.
Built and maintained by Joy Truepath Pte. Ltd., the
team behind the TruePath Office
Mac app (on the Mac App Store). The
engine here (TruePathCore) is the same headless, Foundation-only, zero-
dependency core that ships inside the app — open-sourced for transparency and
standalone use.
xcode-select --install or swift.orggit clone https://github.com/JoyTruepath/truepath-office-mcp.git
cd truepath-office-mcp
swift build -c release
This produces two binaries under .build/release/:
truepath-office-mcp — the MCP stdio servertruepath — a standalone CLI for the same engineGet the server's absolute path for the config snippets below:
echo "$(pwd)/.build/release/truepath-office-mcp"
Use the absolute path printed above wherever you see /path/to/....
Claude Desktop — ~/Library/Application Support/Claude/claude_desktop_config.json
{ "mcpServers": { "truepath-office": { "command": "/path/to/truepath-office-mcp/.build/release/truepath-office-mcp" } } }
Claude Code
claude mcp add truepath-office -- "/path/to/truepath-office-mcp/.build/release/truepath-office-mcp"
Codex — ~/.codex/config.toml
[mcp_servers.truepath-office]
command = "/path/to/truepath-office-mcp/.build/release/truepath-office-mcp"
Cursor / Windsurf — MCP settings (see each IDE's docs for the file location)
{ "mcpServers": { "truepath-office": { "command": "/path/to/truepath-office-mcp/.build/release/truepath-office-mcp" } } }
Restart the host, then ask your agent something like: "Use truepath-office to read the body text of ~/Documents/report.docx."
| Tool | Description | Key parameters |
|---|---|---|
evaluate_formula | Evaluate a spreadsheet formula | formula, data (2D array) |
load_workbook | Load an XLSX file | path |
save_workbook | Save an XLSX file | path, sheets |
query_spreadsheet | Query a cell range | path, sheet?, range? |
load_presentation | Load PPTX slides | path |
save_presentation | Save PPTX slides | path, slides |
load_word_metadata | Load DOCX metadata | path |
read_word_body | Read DOCX body text | path |
extract_docx_structure | Parse DOCX into structured JSON (paragraphs / runs / formatting / tables / sections / headers / footers / footnotes) | path |
pdf_info | Page count & text info | path |
pdf_merge | Merge multiple PDFs | sources, output |
pdf_split | Split a PDF at a page number | path, page |
pdf_protect | Add password protection | path, password |
detect_document_type | Detect type from extension | extension |
list_word_comments | List DOCX comments with resolved state + reply threading | path |
set_word_comment_resolved | Resolve / reopen a comment — whole thread by default (Word-compatible commentsExtended.xml) | path |