spb/anomaly-atlas Public License

Systematic discovery & rigorous validation of statistical anomalies in open HF market data (hfmarketdata.io) — pre-registered, artifact-null-driven, fully reproducible. Live atlas: www.anomaly-atlas.io

quant-research market-microstructure anomalies reproducibility duckdb apple-silicon

Python 61.4% JavaScript 28.7% CSS 8.6% Shell 0.7% Makefile 0.5%

573 B · 18 lines markdown

Rendered Raw Blame History

1---2project: anomaly-atlas3document: Atlas data products index4author: Simon-Pierre Boucher5contact: contact@spboucher.ai6data_source: hfmarketdata.io7created: 2026-08-128status: draft9---1011# Atlas — confidence-labeled findings1213One directory per finding: `<anomaly_id>/<version>/` containing `finding.json`,14`provenance.json` (commit, config, data-manifest hash, hardware, dates) and15`confidence.md` (Level 0-3 with evidence). Entries are created ONLY via16`tools/new_finding.py`, which refuses incomplete provenance/confidence.17Negative results are first-class entries.18