Twinkle Hub

ClaudePartnerby Twinkle AI

Official open data, made queryable. Public records are scattered across hundreds of agencies in inconsistent formats, often only as bulk downloads. We fetch, normalise and index them behind one consistent interface. United States — Federal Register, Congress (bills, votes, members), SEC filings, Federal Reserve data, Census, federal spending, the US Code. United Kingdom — Parliamentary bills, members, food hygiene ratings, national data catalogue. Taiwan — Statute law with full amendment history and the legislature's stated reasoning for each change; court judgments, constitutional interpretations and precedents; parliamentary bills, interpellations, floor votes and attendance; full-text gazette search down to a single motion with its proposers and co-signers; procurement tenders and awards since 2015; real-estate transaction, pre-sale and rental registries. Hong Kong — Legislative Council Hansard and bills, Monetary Authority data, spatial layers. South Korea — Statutes and court precedents. Japan and Singapore — National open data catalogues. Every response states its source and licence. Content we generate ourselves is labelled and never presented as an official record. Coverage is reported honestly: results carry the date range actually held, and when nothing matches we say the query found nothing within the collected range — rather than letting a user conclude the event never happened. Data refreshes daily behind automated quality gates: if an upstream source breaks, returns an implausible row count, or changes structure, the pipeline stops rather than publish bad data. Typical uses: legal and regulatory research, policy analysis, tracking what an elected representative has said and proposed, procurement and property research, fact-checking. This is an independent service, not an official government one. For anything with legal effect, the originating agency's published document governs.