What problem does it solve? Briefs and reports often mix cited facts with reasoned guesses so seamlessly that readers cannot tell which sentences rest on evidence and which rest on priors, leading to over-trusted assumptions or wholesale distrust of the document. This Skill separates the two by tagging every fact-claim inline with its provenance. ## Core Features & Use Cases - Inline provenance tagging: Marks each fact-claim as [validated, {source}] when a specific source states it, or [assumption, retire-by {milestone}] when it is reasoned from priors, with every assumption tied to a concrete event, date, or threshold that will test it. - Coverage reporting: Emits a coverage line verifying that at least 80% of quantitative sentences, 100% of headline claims, and 100% of forward-looking projections are tagged. - Vendor self-report handling: Treats a party's claims about itself as assumptions requiring independent verification rather than validated facts. - Use Case: A market brief claims "$6.25B market size" and "24.8% CAGR" alongside a projection that "skills graphs will dominate by 2027". The Skill tags the first two with their analyst-report citations and the projection as an assumption with a dated retire-by milestone, so a decision-maker sees exactly which numbers are sourced. ## Quick Start Go through this draft and tag each claim as validated with its source or as an assumption with a specific retire-by milestone, then show me the coverage summary.
