What problem does it solve? Before citing or sharing content from an unfamiliar website or a viral post, you need to know whether the source and its claim can be trusted — and reading the page itself cannot answer that. This Skill applies Mike Caulfield's SIFT method (2019), built on lateral-reading research, to vet a source or claim by checking what the wider web says about it, returning a clear verdict of Trusted, Usable with caveat, Unconfirmed, or Refuted. ## Core Features & Use Cases - Four-move verification workflow: Runs Stop, Investigate the source, Find better coverage, and Trace claims to the original context, all performed laterally on sites other than the one being checked. - Structured verdict output: Produces a copy-ready report naming the claim, the evidence found at each move, one of four verdicts, and the move that settled it. - Sibling-skill hand-offs: Routes results to source grading, corroboration, or refusal workflows instead of leaving claims unresolved. - Use Case: A screenshot claiming a chip is "100× faster than GPUs" is circulating. The Skill traces it to a narrow vendor microbenchmark, finds no independent coverage of the general claim, and returns Unconfirmed with a refusal hand-off. ## Quick Start Run a SIFT check on this link before I cite it — it is a site I have never heard of claiming a new battery chemistry doubles EV range.
