LabscoMCP early access ↗
BrianCusackMCP SERVER

Sharepoint

Reach into an organisation's SharePoint drive — search folders, list items, download and read files.

by BrianCusack ↗Office Suites & Cloud File StorageCommunity

LABSCO SUMMARY

Read-side access to a SharePoint drive, through your own app registration.

It is the narrow useful slice: find the folder, see what is in it, and pull a document back into the conversation — with the credentials scoped by the app registration you created rather than by a shared login.

What it is

A connector to organisational SharePoint through the Microsoft Graph credentials of an app registration. The repository's own implementation table is the honest map of it: sites, drives, lists and calendar are all sketched out, and a handful of the drive operations are the ones actually wired up.

What you get

  • Search folders across the drive, and download a file
  • Site listing and file reading exposed as resources, with drive items available as a dynamic resource
  • Three prompts the client can call directly: document-summary, find-relevant-documents, explore-folder

Requirements

An Azure app registration with access to the site: TENANT_ID, CLIENT_ID and CLIENT_SECRET, plus the SITE_ID and DRIVE_ID you are pointing at — copy .env.example to .env and fill them in. Build with pnpm run build and run node, or docker build -t mcp/sharepoint . and pass the five values as -e flags. The README labels the project work in progress and for research at this stage, and its table marks the write operations — create folder, upload, move, copy, list and calendar CRUD — as not implemented.

Setup effort

Run a container — build the image from the repository, then point your client at it