Connecting Integrations

Integrations overview — the catalogue

Navigate to Integrations in the sidebar to see all available data sources and tools that can be connected to your AI assistant.

The catalogue

The page shows two groups of integration cards:

• Installed integrations — sources you have already connected. These appear at the top.
• Available integrations — sources that are available but not yet set up.

Each card shows the integration name, description, and category tags (e.g., Knowledge Base, Tool, Communication Channel).

Searching and filtering

• Search bar — type any part of an integration's name or description to filter the list instantly.
• Tag chips — click a category tag (Knowledge Base, Tool, etc.) to show only integrations in that category. Click it again to deselect.

Connecting a new integration

Admin only. Only users with the admin role can add new connections.

1. Find the integration you want in the available list.
2. Click its card. You are taken to that integration's detail page.
3. Follow the on-screen instructions to authenticate and configure the connection.

Connections you set up appear under Installed integrations on future visits.

Website integration — crawling your site

The Website integration lets ragistry automatically crawl and index your website so the AI assistant can answer questions using your published content.

Adding a website connection

1. Open Integrations and click the Website card (or navigate directly via the Websites shortcut).
2. On the integration detail page, click + New Connection.
3. Enter the full URL of your website (e.g. https://www.example.com) and save.
4. The crawler starts automatically.

Up to 3 connections (websites) can be active at one time.

Monitoring the crawl

The connection detail page shows:

Indicator	Meaning
Status card	Current state: Crawling, Connected, Pending, Error
Pages card	Total pages found; indexed pages shown as a note
Progress bar	% of URLs in the queue that have been processed
Pages per hour	Crawl speed (shown while crawling is active)

The page updates automatically via real-time subscription while the crawler is running — no need to refresh manually.

Stopping the crawler

If you need to stop a long-running crawl, click the Stop button that appears in the page header while crawling is active. The crawler will finish its current queue batch then halt.

Re-syncing

Once a crawl is complete, click Synchronise to trigger a fresh crawl. This picks up new and changed pages.

Managing indexed pages

The Crawled Pages table at the bottom shows every indexed URL. You can select rows and delete individual pages from the index using the delete action.