An open-source automated platform that scores U.S. political events on two independent axes — constitutional damage and media distraction — publishing immutable weekly snapshots with full algorithmic transparency.

The Distraction Index is a civic intelligence platform that uses AI (Claude API) to score every major U.S. political event on two dimensions: how much real institutional damage it causes (Score A) and how much media attention it generates (Score B). Events where damage is high but attention is low are flagged as "undercovered." Events where distractions coincide with undercovered damage are detected as "smokescreens."

The platform publishes immutable weekly snapshots — once a week closes, scores are permanently frozen. All scoring formulas, weights, and AI prompt templates are documented publicly. The entire codebase is open source.

Began December 2024. After 59+ weeks: 1,500+ scored events, 11,800+ ingested articles, 210+ smokescreen pairs identified. Built with Next.js, Supabase, Claude API, and deployed on Vercel.

Target users: citizens seeking media literacy tools, journalists covering democratic institutions, researchers studying agenda-setting and media framing, civic technologists building transparency tools.