Skip to main content
SupaCortex
Sign in Book a Demo
Feature

The knowledge already sitting in your drives and wikis

Policies in SharePoint, runbooks in Confluence, the price list somebody keeps in Google Drive. Connect the account once, pick what to read, and SupaCortex keeps a project in step with it.

Book a demo See pricing
  • OAuth or service account
  • Folder, site and space pickers
  • Only changed files fetched

How do the Google Drive, SharePoint and Confluence connectors work?

An administrator connects the account once for the whole workspace - by signing in with OAuth, or with a service account, an app-only registration or an API token. In any project, a feed then picks that connection and the folders, document libraries or spaces to read. The first run reads everything in scope; later runs ask the service what changed and fetch only those items, and deletions at the source are removed from the project.

  • Google Drive: folders or a shared drive; Docs, Sheets, PDF and Office files.
  • SharePoint: document libraries, optionally site pages, through Microsoft Graph.
  • Confluence Cloud: spaces, optionally filtered by label.
  • Read-only access; tokens and keys encrypted and never shown again.
  • Incremental runs fetch only what changed since the last one.
  • Available on Growth, Business and Enterprise.

Google Drive

Beta

Sign in with Google, or share folders with a service account. Docs come in as text with their headings; Sheets as spreadsheets; PDFs and Office files as they are.

SharePoint

Beta

Sign in with Microsoft, or use your own Entra app with read-only site access. Pick a site, then the libraries - and site pages too, if you want them.

Confluence

Beta

Sign in with Atlassian, or use an API token. Pick spaces; pages keep their headings, tables and code blocks.

How it works

Connect once, feed many projects

  1. 01

    Connect the account

    An administrator connects Google, Microsoft or Atlassian in workspace settings. One connection can serve feeds in any project.

  2. 02

    Pick what to read

    The feed form browses the account: folders and shared drives, sites and libraries, or spaces.

  3. 03

    Label the feed

    Set the category, tags, trust tier and, if it is one customer's material, a customer ID - once, for every file.

  4. 04

    Stay in step

    Each scheduled run fetches only what changed. Deleted files are removed; Run now refreshes on demand.

In detail

What happens behind the picker

Read-only, and secrets stay secret

A knowledge feed has no business writing to your drive, so it never asks to.

  • Read-only scopes: Drive read-only, SharePoint and files read, Confluence content read.
  • Refresh tokens, service-account keys, client secrets and API tokens are encrypted at rest and never returned.
  • If a connection is revoked, it is marked "needs re-auth" and the feed stops - nothing is deleted.
  • A connection in use by a feed cannot be removed until the feed is.

Incremental, so large libraries stay cheap

Re-reading ten thousand files to find the three that changed wastes everyone's time.

  • Drive's change list, Microsoft Graph delta and Confluence's last-modified search find what changed.
  • Only those files are downloaded and re-indexed; the rest are left alone.
  • Drive and Confluence also run a full listing every few runs, so nothing removed is missed.
  • A per-feed document limit keeps the first run bounded.

Related

Internal Knowledge Assistant

One place your whole team asks "how do we...?" - across policies, wikis, drives and past tickets.

Team Chat Over All Knowledge

Your team asks across every project at once, with a private history and sources on every answer.

Knowledge Projects

Keep each product, team or audience in its own project - its own sources, widget and API keys.

Website & Help-Centre Crawling

Point it at a help centre, Google Site or docs site and it stays in step on a schedule.

Integrations

Every source it learns from and every place it answers, with what is live and what is next.

Security

Separate databases per workspace, encrypted secrets, and answers scoped to who is asking.
Frequently asked

Questions people ask before they start

No - and it is worth being clear about. SupaCortex does not copy per-user sharing from the source. Everything a feed reads goes into one project, and anyone who can ask that project can be answered from it. You choose what each project contains and who can ask it, so point feeds only at folders, sites and spaces that suit the project's whole audience.

No. You can sign in with OAuth, or use a non-interactive connection instead: a Google service account that you share folders with, a Microsoft Entra app with read-only site permission, or a Confluence API token.

Google Drive: Docs, Sheets, PDFs, Word, Excel, Markdown and text files in the folders or shared drive you pick. SharePoint: files in the document libraries you pick, and optionally site pages. Confluence Cloud: pages in the spaces you pick, optionally filtered by label.

The connectors are built and tested against recorded provider responses, and we want a few more live workspaces on them before we drop the label. If you want them in a trial, we set them up with you.

Growth, Business and Enterprise. Each feed counts towards the plan's feed limit, and the fastest schedule is every six hours on Growth and hourly on Business and Enterprise.

Pick one folder your team asks about every week. Start there.

Book a 30-minute demo on your own documents. We set up a 14-day trial workspace for you - no credit card, no setup fee.

Book a demo See pricing

Questions first? Email info@greatdevelopers.com or call +91 8585953085.

Still curious? Ask your favorite AI about SupaCortex.

ChatGPT Claude Perplexity Grok