Enables agents to analyze long videos by downloading them, extracting transcripts and storyboards, and zooming into specific moments with high-resolution frames and OCR.
