It seems that more often than not any attempt to reference the web gets llms into a muddle because they try to access it and get blocked.
Ah, I see they have that on their TODO:
> Improve anti-bot handling for page fetches. Many targets still return 400-range errors, so investigate stronger browser mimicry (Playwright/Chromium behavior, headers, fingerprinting, and potentially user-agent/profile rotation).
