Introduced a new tool for extracting YouTube video transcripts and leveraging them to answer questions. Updated `SearchAndRead` to support reading YouTube transcripts and regular pages distinctly. Included relevant dependencies for handling subtitles and video downloads.
Removed redundant fields and callbacks in the search agent while introducing concurrent processing for reading search results. Updated logic to enhance readability and modularity, including capped reads and streamlined interaction with search results. Adjusted dependencies and related usage to align with the refactored design.
Introduce multiple agents, tools, and utilities for processing, extracting, and answering user-provided questions using LLMs and external data. Key features include knowledge processing, question splitting, search term generation, and contextual knowledge handling.
