Returns one entry per profile with its friendly ``name``, on-disk
``directory`` id, and whether it is the default profile. Use the ``name``
(or ``directory``) value as the ``profile`` argument to ``get_history``.

A whole day can hold thousands of visits, so the result is paginated and the
day can be narrowed to a time window.

Args:
    profile: Profile friendly name (from ``list_profiles``) or directory id.
    date: Day to fetch, formatted ``YYYY-MM-DD``. Day boundaries are
        interpreted in the local machine timezone.
    start_time: Optional lower bound within the day, ``HH:MM`` (24-hour,
        local time). Defaults to the start of the day (``00:00``).
    end_time: Optional upper bound within the day, ``HH:MM`` (exclusive,
        local time). Defaults to the end of the day. Must be later than
        ``start_time``.
    limit: Maximum number of entries to return per page (default 10000,
        capped at 50000). Non-positive values fall back to the default.
    offset: Number of entries to skip from the start of the window, for
        paging through large results (default 0).

Returns:
    A dict with:

    * ``entries``: list of page visits in the window, ordered by visit time
      ascending. Each entry has the local ISO ``visit_time``, ``url``,
      ``title``, ``visit_count``, ``typed_count``, ``transition`` type, and
      the ``url_id``/``visit_id`` identifiers.
    * ``offset``/``limit``: the paging window that was applied.
    * ``has_more``: whether more entries follow this page.
    * ``next_offset``: the ``offset`` to pass to fetch the next page, or
      ``null`` when ``has_more`` is false.

Prefer this tool over ``get_history`` for questions like "what did I browse
today, hour by hour?": it returns far less data and never exposes full URLs,
which can embed auth tokens, session ids or other secrets. It does include a
small sample of page titles per domain for context. Only use ``get_history``
when specific URLs are genuinely required.

Only ``http``/``https`` visits are aggregated; ``file:``, ``edge:``,
``chrome:``, extension and other non-web URLs are skipped (and counted).

Args:
    profile: Profile friendly name (from ``list_profiles``) or directory id.
    date: Day to summarize, formatted ``YYYY-MM-DD``. Day boundaries are
        interpreted in the local machine timezone.
    start_time: Optional lower bound within the day, ``HH:MM`` (24-hour,
        local time). Defaults to the start of the day (``00:00``).
    end_time: Optional upper bound within the day, ``HH:MM`` (exclusive,
        local time). Defaults to the end of the day. Must be later than
        ``start_time``.
    group_by: Bucket granularity. Currently only ``"hour"`` is supported.
    limit: Maximum number of underlying visit rows to aggregate per page
        (default 10000, capped at 50000). Non-positive values fall back to
        the default.
    offset: Number of visit rows to skip from the start of the window, for
        paging through large days (default 0).
    include_titles: Include a small sample of page titles per domain
        (default true). Set false to get domains and counts only.
    max_titles_per_site: Maximum sample titles per domain per bucket
        (default 3, capped at 10). Values <= 0 return no titles.

Returns:
    A dict with:

    * ``profile``/``date``/``group_by``/``timezone``: the query context.
    * ``entries_considered``: visit rows examined in this page.
    * ``web_entries``: rows counted in the buckets.
    * ``non_web_entries_skipped``: rows skipped as non-web URLs.
    * ``buckets``: hour buckets ordered ascending, each with the local ISO
      ``hour``, a friendly ``label`` (e.g. ``8 AM``), ``total_web_visits``
      and ``sites`` (``domain`` + ``visits`` + ``sample_titles``, sorted by
      visits descending then domain ascending). Domains are normalized
      (``www.`` stripped); no full URLs, query strings or fragments are
      returned. ``sample_titles`` holds up to ``max_titles_per_site``
      deduplicated, truncated page titles in first-seen order.
    * ``offset``/``limit``/``has_more``/``next_offset``: paging metadata,
      with the same semantics as ``get_history``.