All sources

Website

Index a documentation site by its address. Public, or behind a token, password or API-key header.

What you will be asked for

  • Start URL
  • Stay under
  • Page limit
  • Authentication
  • TokenSecret
  • Username
  • PasswordSecret
  • Header name
  • Header valueSecret
  • Obey robots.txt

Where to find these

Point Worclaude at documentation that lives on the web and it will crawl, chunk and index it.

Start URL

The page to begin from — usually a docs index:

arduino
https://docs.example.com/

Path prefix

Keeps the crawler inside one section. With a prefix of /guides, a link to /blog/hello is ignored. Leave blank to allow anything on the same host.

The crawler never leaves the starting host regardless.

Max pages

A ceiling, so a mistyped prefix cannot crawl for hours. A few hundred covers most documentation sites. Start low; re-syncing with a higher number is cheap.

Authentication

Only if the site is not public.

  • Bearer token — paste the token; sent as Authorization: Bearer ….
  • Basic — a username and password.

If the site is behind an SSO login page rather than HTTP auth, neither will work: the crawler will index the login page. Publish the docs internally or use a different source.

Connect this source

About a minute, once you have the credentials.