Web Search

Web Search gives Casper access to current information from across the web instead of relying only on knowledge already available to the model. It can search for relevant sources, combine and prioritise results, open promising pages, extract readable content, and use that information to answer questions with current context. It can: Search the live web for current information Return multiple relevant sources Remove duplicate results Prioritise results likely to contain the answer Read the full content of useful search results Open a direct URL without performing a search first Extract readable content from webpages Extract text from PDF documents Use deeper page content instead of relying only on search snippets Surface useful contact details such as email addresses and phone numbers when found Fall back to directly reading thin search results when more context is needed Return source titles and URLs alongside retrieved information The tool is specifically designed to go beyond basic search-result snippets by retrieving richer source content where available and enriching weaker results when necessary.