Data Licensing & Architecture

What Chart Library Actually Serves

Chart Library is a derived-data product. The public API exposes outputs computed from raw market data, not the raw data itself:

• 384-dimensional pattern embeddings (compressed vector representations of chart shape)
• Pattern match rankings with L2 distance scores
• Forward return aggregates: means, medians, win rates, percentile ranges
• AI-generated plain-English summaries
• Regime classifications (bull+calm, bear+volatile, etc.)
• Context similarity scores across 12 dimensions (VIX, yield curve, breadth, etc.)

These outputs cannot be reverse-engineered to reconstruct the underlying OHLCV data — a requirement under all major exchange "derived data" license frameworks (CME, NYSE, NASDAQ).

What We Do Not Redistribute

Our public API does not serve raw minute bars, last-sale prices, real-time quotes, order book depth, trades, or other raw market data. The candlestick charts rendered on chartlibrary.io qualify as "display" under upstream data license terms — they are shown to our users on our own property, not redistributed via a public programmatic API.

Data Sources

Polygon.io / Massive

Historical and end-of-day US equity OHLCV bars, corporate actions, and ticker reference data. Powers our 24M+ pattern embedding corpus.

FRED (Federal Reserve Economic Data)

Treasury yields, credit spreads, and macro indicators. Free and publicly redistributable.

SEC EDGAR

Corporate filings — 8-K material events, 10-Q/10-K periodic reports, and earnings dates. Public domain, commercial use permitted.

Financial Modeling Prep

Earnings calendar (upcoming reports). Commercial tier.

Internal computation

All pattern embeddings, forward returns, regime classifications, and AI summaries are computed by Chart Library from the raw data above. These derived works are the core product.

API Developers: What You Can Redistribute

If you build on the Chart Library API, you can redistribute our derived outputs (pattern matches, forward returns, AI summaries, regime labels, context scores) under the terms of your API plan.

You cannot attempt to reconstruct raw market data from our outputs or use the API to build a raw data redistribution service. Attribution ("Powered by Chart Library") is required on public-facing integrations.

Accuracy and Disclaimers

Historical pattern matching is not predictive of future results. Past performance does not guarantee future returns. Our outputs are historical facts about what happened after similar charts, not financial advice. See our full disclaimer.