Why ProfileScribe keeps fact extraction separate from published claims
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, and those facts sit in a review queue before anything becomes a claim someone else can read.
That separation exists because scattered professional work (repos, links, writing) is easy to extract but easy to overstate. Keeping extraction and publishing as distinct steps means a profile only says what the underlying evidence actually supports, and changes stay traceable back to a source instead of blending into an unverifiable summary.
It's a small structural decision, but it's the one the rest of the system depends on.