Finalist
FenceFlow
FenceFlow eliminates paper-based workflows from fencing tournaments using Opus 4.6.
Core trick
Vision OCR with extended thinking extracts scores from photographed handwritten sheets, while an agent auto-approves clean sheets and flags anomalies for review.
Description
FenceFlow eliminates paper-based workflows from fencing tournaments using Opus 4.6. Built from hosting 30+ events, it solves a real problem: staff squint at illegible handwriting, results lag hours, and errors change seedings. Four interfaces: The Committee Dashboard runs an autonomous agent that auto-approves scoresheets, flags anomalies for human review, and logs decisions with reasoning. The Referee Portal uses Vision OCR with extended thinking to extract scores from photographed handwritten sheets. The Coach View runs a Bradley-Terry model with Monte Carlo simulations and multi-turn Opus chat for advice. The Public View streams live brackets and AI narrator commentary token by token. Fencing software lacks public APIs so the demo uses data from a tournament the developer hosted.
README excerpt
# FenceFlow — AI-Powered Tournament Operations **Built with Anthropic Opus 4.6** | by CozMx > *Build a Tool That Should Exist — Eliminate busywork. Make hard things effortless.*  ## Demo [](https://www.youtube.com/watch?v=Yai0l43kM9c) ## The Problem Running a fencing tournament is a logistical nightmare. Score sheets are handwritten, results must be manually entered from photos, referees need constant coordination, and spectators have no visibility into live results. A single tournament with 120+ fencers generates hundreds of score sheets, announcements, and bracket updates — all managed by a small volunteer committee under time pressure. FenceFlow replaces this manual chaos with an AI-powered operations platform that automates scoring, coordinates referees, and delivers real-time results to every stakeholder. ## 5 AI Features ### 1. Vision OCR with Extended Thinking Referees photograph handwritten score sheets. A first-pass Sonnet model reads the scores via OCR. When confidence drops below 80%, Opus 4.6 activates **extended thinking** — a second-pass analysis that re-examines ambiguous cells, applies fencing rule constraints (V0-V5 range, diagonal symmetry), and produces a transparent reasoning chain. The committee sees exactly which cells were corrected and why. ### 2. Autonomous Tournament Agent An agentic loop powered by Opus 4.6 with **tool use** monitors the to