Compare
Chart Library vs StockCharts — the chart versus what the chart did next.
Not competitors. StockCharts is a long-running charting and technical-analysis platform — indicators, overlays, annotations, scans, and SharpCharts — for a human to look at a chart and apply their own rules.
Chart Library doesn’t draw charts. It takes a real (symbol, date, timeframe) and returns the cohort of historical days whose shape most resembles it, what they did over the next 1/5/10 days, and a calibrated outcome distribution with an audit receipt — served to an agent over API/MCP. Indicators are rules a person reads; Chart Library is the empirical record of what that shape’s history did.
Look at the chart vs quantify its precedent
- StockCharts = visualization & TA for humans. Indicators and annotations you interpret yourself.
- Chart Library = calibrated analog outcomes. The cohort’s forward-return distribution + drivers + receipt (nominal 80% band held 80.8% across 300K+ audited cases). No indicator to configure — it matches raw shape.
- Built for the agent, not the eyeball. An MCP/REST endpoint returning structured, calibrated answers — never a directional call.
Frequently asked questions
- Is Chart Library a StockCharts alternative?
- Not really — it's a different layer. StockCharts is charting/TA you look at; Chart Library returns the calibrated historical-analog outcomes for a setup over API/MCP, for an agent. Use a charting tool to view, Chart Library to quantify what similar charts did.
- Does it have indicators, overlays, or scans?
- No. It matches raw price/volume shape (no indicator config) and returns the analog cohort's calibrated outcome distribution. For visual TA, use StockCharts; for the base rates, call Chart Library.
- Does it predict the chart's next move?
- Never. Similarity-only by design — it surfaces what analogs did next as a calibrated distribution with a receipt, no forecast.
Try it
Run a cohort_analyze call.
Free Sandbox tier — 1,000 calls/day, no authentication. MCP install for Claude or Cursor takes 30 seconds.