The audit log is the browsing history

Your agent fleet visits hundreds of vendor pages a day. When something goes wrong: a wrong invoice paid, a contract canceled by mistake: the question is always the same: where did the agent actually go.

GET /history on a Hoody Browser instance returns every navigation, persistent across restarts, with timestamps, HTTP status, the requesting tab, the originating session, and a source field that distinguishes API-driven navigations from manual ones. You filter by domain=vendor.com&since=2026-04-15T00:00:00Z and get the agent's full trail to that domain in one response.

You don't need a separate audit pipeline. You don't tail logs. The browser already remembered, and it remembered with structure.

The agent's browsing history is the audit log; the audit log is a query.