Skip to main content

​
How it works

The MediaWiki connector indexes pages from any public MediaWiki site, for example a company wiki or a Fandom wiki. You choose the pages to index by category, by page name, or both. Each wiki page becomes one document. The page categories are stored as document metadata. The connector reads the wiki through its public API. It does not need a credential. On each poll, the connector re-indexes pages that changed since the last run.
To index Wikipedia, use the Wikipedia connector instead.

​
Setting up

1

Enter the site details

Go to Admin Panel → Connectors and select MediaWiki.
  • Language Code: the language code of the wiki, for example en. The default is en.
  • MediaWiki Site URL: the URL of the wiki, for example https://wiki.example.com.
2

Choose what to index

  • Categories to Index: category names to index, with or without the Category: prefix. The connector indexes every page in each category.
  • Pages: names of individual pages to index.
  • Recursion Depth: how many levels of subcategories to follow. 0 indexes only the listed categories. -1 follows every subcategory.
Enter at least one category or page.
3

Choose document access

Under Document Access, select Everyone in Your Organization or Specific Groups. Specific Groups needs the Business or Enterprise tier on Onyx Cloud, or the Enterprise Edition when self-hosted.
4

Start indexing

Select Create Connector to start indexing.
The MediaWiki connector does not support Sync with MediaWiki (permission sync).