Glossary · Data & retrieval

Deep search

A search that does not stop at the ranking — it scrapes the top results for content, as a background job.

What is deep search?

Deep search runs a query, then reads the top matches platform-aware in a real browser and returns cleaned content. Because that read is real browser work rather than a fetch, it is asynchronous: the call returns a job id and the result is collected by polling or a webhook.

Why it matters

A ranking tells an agent where to look; it does not tell it what is there. Deep search closes that gap in one call, and does it asynchronously so a multi-page read does not have to fit inside a request timeout.

Related terms

See it in context

Back to the full glossary

Put your agent to work. Keep the decision.

Give your agent a real browser, authorize the accounts it may use, and keep control of every step. Repeated runs compile into skills that replay without a model.