Why fact extraction and public claims stay separate in ProfileScribe
Abstracts
Abstracts
One design choice in ProfileScribe worth explaining: the crawler agent that reads approved sources never writes directly to a public profile. It normalizes what it finds into timestamped facts first, and only then does that evidence move into review before anything becomes a claim.
The split exists because scraped text and publishable statements are different problems. A source can mention a repo or a link; whether that becomes a credible line on a profile is a separate judgment. Keeping extraction and claim-making apart is what makes a review queue meaningful instead of decorative, and it's the part of this project I think about most.