Prerequisites
- A Confluence Cloud site (e.g.
acme.atlassian.net). - An Atlassian account with read access to the spaces you want indexed — a dedicated service/bot account is recommended so the connector doesn’t break when a teammate leaves.
- A BeforeQuery workspace
ownerrole.
Set up the connector
1
Create an Atlassian API token
Sign in as the account the connector should use and create an API token at id.atlassian.com → Security → API tokens. Copy it — you’ll need the token and the account’s email address.
2
Add the source in BeforeQuery
In the dashboard, open your Project → Sources, click Add source, and choose Confluence. Fill in:
- Site URL — your Confluence base URL, e.g.
https://acme.atlassian.net - Email — the Atlassian account email
- API token — the token from step 1
- Space key (optional) — index a single space (e.g.
DOCS); leave empty to index every space the account can read
3
Start the first sync
Click Add. Pages are fetched, converted to markdown, chunked, and embedded. Track progress on the source card and browse the results under Documents.
Connection settings
Keeping content in sync
Use the Sync button for on-demand refreshes, or set a daily/weekly schedule on the source. Only changed pages are re-embedded.Troubleshooting
- A space isn’t being indexed? The connector account can’t read it — grant the account access in Confluence space permissions, then sync again.
- 401/403 errors during sync? The API token expired or was revoked. Create a new token and update the source settings.
- Too much indexed? Set a space key to limit indexing to one space, or split spaces across multiple Confluence sources and organize them with source groups.
- Keeping wiki content internal? Mark the documents
restrictedso they answer only for workspace members, never for public widget users.