Pine Quant Analysis Skill

Create, run, and visualize technical indicators anywhere Using Pine Script.
This skill teaches LLM agents and developers how to:

• Write native Pine Script indicators
• Execute them on historical or live data using PineTS
• Extract computed indicator values programmatically
• Visualize price action and indicators with QFChart

Designed for research agents, trading bots, dashboards, and quantitative systems.

---

What This Skill Unlocks

• Run Pine Script logic in Node.js, browsers, Deno, Bun
• Compute indicators on any OHLCV data source
• Separate indicator logic from visualization
• Automate backtests, analytics, alerts, and charts

---

Prerequisites

• JavaScript or TypeScript runtime
• OHLCV market data (exchange API, CSV, database, or stream)
• PineTS and QFChart libraries

npm install pinets @qfo/qfchart

---

Step 1 - Define a Technical Indicator (Pine Script)

PineTS executes real Pine Script (v6) exactly as written.

Example: EMA Crossover Indicator

const indicatorScript = `
//@version=5
indicator("EMA Cross", overlay=true)
fast = ta.ema(close, 9)
slow = ta.ema(close, 21)
plot(fast, "Fast EMA")
plot(slow, "Slow EMA")
`;

This indicator computes two exponential moving averages and plots them on price.

---

Step 2 - Run the Indicator with PineTS

Create a PineTS engine, attach a data provider, and execute the script.

import { PineTS, Provider } from 'pinets';

const engine = new PineTS(
  Provider.Binance,
  "BTCUSDT",
  "1h",
  200
);

const { marketData, plots } = await engine.run(indicatorScript);

What You Get

• marketData → OHLCV time series
• plots → indicator outputs indexed by plot name

plots["Fast EMA"].data;
plots["Slow EMA"].data;

Each plot is a time-aligned numeric series ready for analysis or visualization.

---

Step 3 - Use Indicator Data Programmatically

Indicator outputs can power:

• Signal generation
• Backtesting logic
• Alerts & notifications
• Machine learning pipelines

const fast = plots["Fast EMA"].data;
const slow = plots["Slow EMA"].data;

const crossover = fast.at(-1) > slow.at(-1);

No charts required - indicators are just data.

---

Step 4 - Visualize with QFChart (Optional)

QFChart renders high-performance financial charts with multiple panes and overlays.

Format Market Data

const candles = marketData.map(d => ({
  time: d.time,
  open: d.open,
  high: d.high,
  low: d.low,
  close: d.close,
  volume: d.volume
}));

---

Render Price + Indicators

const chart = new QFChart.QFChart(document.getElementById("chart"), {
  title: "BTC/USDT - EMA Crossover",
  height: "600px"
});

chart.setMarketData(candles);
chart.addIndicator("Fast EMA", plots["Fast EMA"].data, { isOverlay: true });
chart.addIndicator("Slow EMA", plots["Slow EMA"].data, { isOverlay: true });

• Overlay indicators on price
• Or render them in separate panes (RSI, MACD, etc.)

---

Step 5 - Real-Time & Streaming Updates

Both PineTS and QFChart support incremental updates.

chart.updateIndicator("Fast EMA", newFastValues);

Use this for live dashboards, bots, or monitoring agents.

---

Common Indicator Patterns

• Trend indicators: EMA, SMA, VWAP
• Momentum: RSI, Stochastic, MACD
• Volatility: Bollinger Bands, ATR
• Custom signals using full Pine Script logic

PineTS supports 60+ built-in Pine Script TA functions.

---

Best Practices for Agents

• Treat Pine Script as pure computation
• Keep visualization optional
• Cache indicator outputs for reuse
• Run multiple timeframes via multiple engines

This enables scalable, composable quantitative systems.

---

Quick Reference

# Execute indicator
engine.run(pineScript)

# Access indicator values
plots["RSI"].data

# Visualize (optional)
chart.addIndicator(...)

---

Learn More

• PineTS - Pine Script runtime & engine
• QFChart - Financial charting library
• QuantForge - Quantitative tooling ecosystem

---

Don't forget to star PineTS and QFChart projects if you like them : https://github.com/QuantForgeOrg/PineTS , https://github.com/QuantForgeOrg/QFChart