Browser Reader
Fetches a web page and reads it as text — a documentation page, a changelog, a competitor's pricing.
Where web search finds the page, the browser reader opens it. Give an agent a URL and it comes back with the readable content rather than a snippet.
Size and time are capped, redirects are re-checked against the same address policy as the HTTP tool, and the page's text is fenced as untrusted before it reaches the model.