Here's what it does: - Natural-language search over the full corpus - Results are document discovery, not “trust me” summaries - Every result includes clickable citations that jump to the exact source page in the committee’s Google Drive so you can verify context quickly

Some useful test queries: - “Find documents mentioning [person/org] in connection with flights / schedules / contacts” - “Show mentions of ‘massage’, ‘modeling’, ‘recruiting’, ‘Palm Beach’, ‘New York’, ‘Little St. James’” - “What documents reference [date range] and [location]”

Known limitations: - OCR noise on low-quality scans - Names/aliases can be inconsistent; citations are the ground truth

I'd love to pose some quick questions (If I had more time) that collate the relevant data from the files, such as the contents linking DJT, or any other high profile individual.