MBio AI Chat Docs
Documentation / AI & RAG

Website crawler & knowledge sync

Queue safe same-host crawls that discover pages, extract readable text, create chunks and build embeddings for the selected website.

Create a crawl job

Open Crawler, enter a public HTTP/HTTPS URL, select Page, Website or Sitemap mode, then choose a page limit. Jobs run through the crawler worker so the Admin request stays fast.

Run the worker

php bin/process-crawls.php 25

Use Cron or a process supervisor in production. Review job status, discovered pages and indexed pages from the Crawler page.

Isolation & safety

The crawler rejects localhost and private/reserved network targets, limits response size and indexes content only into the selected website knowledge base.