Confluence

Deskhand reads Confluence Cloud, not Server or Data Center, through the REST API v2. It signs in with CONFLUENCE_EMAIL and CONFLUENCE_API_TOKEN, and the account needs read access to the pages you index.

1. Create an API token

Create a token at id.atlassian.com for an account that can read the pages you want indexed.

Anyone who can message Deskhand can read, through it, any document its docs credentials can see within the configured scope. Use a low-privilege account, scoped to what everyone may read.

CONFLUENCE_EMAIL=[email protected]
CONFLUENCE_API_TOKEN=...

2. Add the source

docs:
  sources:
    - type: confluence
      name: wiki
      site: https://acme.atlassian.net
      space_key: IT
      parent_page_id: "123456"

Key	Meaning
`site`	Your Atlassian site URL.
`space_key`	The space to index, such as `IT`.
`parent_page_id`	Only this page and the pages under it. Quote it.

Set space_key, parent_page_id, or both.

Scope

With only space_key, Deskhand indexes every page in the space.
With parent_page_id, it indexes that page and its descendants. The ID is in the page URL: https://acme.atlassian.net/wiki/spaces/IT/pages/123456/VPN+setup.
With only parent_page_id, Deskhand finds the space from the page.

Pages are converted to Markdown for search, and answers link to the page in Confluence. Deskhand syncs on startup and every docs.sync_interval_minutes, re-fetching only pages that changed.