Help Center / Data & Sources

About the Library

The Library is the evidentiary foundation of Disclosure Navigator. Everything the platform knows — every entity, event, relationship, and briefing answer — is grounded in documents and media that have been ingested, processed, and made searchable here.

What the Library Contains

The library currently includes the following collections:

Additional government and researcher collections are ingested on an ongoing basis. The SOURCES tab always reflects the current state of the library.

How Documents Are Processed

Every document follows the same ingestion pipeline. PDFs are ingested and full text is extracted. Video content is transcribed. The extracted text then goes through a multi-pass AI analysis that identifies and extracts:

All extracted data is linked back to the source document and the specific passage it came from. Nothing in the platform is asserted without a traceable source.

Entity and Event Extraction

Every entity profile in the Explorer was identified in at least one source document — entities are not created speculatively. Relationship edges between entities are drawn from documented co-mentions and explicit statements in source material, not from inference or general world knowledge. The same applies to events: if it appears in the Timeline, a source document says it happened.

Library Growth

New document collections are reviewed and added regularly. Priority is given to primary government sources, then to high-quality research sources with documented methodologies and named witnesses. The SOURCES page lists every active collection with document counts and collection metadata.

Want to suggest a source? If you believe a specific document collection, government release, or researcher archive should be in the library, contact us. We evaluate every suggestion against our source quality standards before ingesting.