Skip to main content
Larkup gives you one Data page for every source. Add content once, and Larkup indexes it in the background.
Data page with Files, Text, Website, Media, and Integrations tabs

The Data page, with a tab for every source

Search by keyword or paste a direct URL. The built in local crawler works immediately; switch to the Cloud web crawler in Settings for JavaScript heavy or protected sites.
Website search results ready to add

Starting a crawl by keyword

Jobs run in the background. Track pages processed and status right in the tab.
Running crawl job with page count and status

A running crawl job

Your corpus

Every loaded document lands in Data → All Data, with its source, status, and character count.
All Data table with indexed documents

All Data, every indexed document in one table

Organize sources with groups

Open Data → Groups to create collections and control whether each collection is available to the Assistant. Use a custom group’s actions menu to rename it or delete it. Only empty custom groups can be deleted. Move their sources to another group from All Data, or delete those sources first. The Default group cannot be deleted.

Scraping settings

Configure the web scraper and web search provider under Settings → Search & Scraping. No .env editing required.
Web search and web scraper provider settings

Search & Scraping settings

The built-in crawler works through the Larkup API without Docker, a separate browser service, or an API key. For JavaScript-heavy or bot-protected sites, you can optionally select Firecrawl Cloud and add its API key. Organizations that route traffic through a proxy can configure it here too.
Proxy server, username, and password fields

Proxy settings for secure scraping