indicator-expert（指标专家）

OpenAlgo Indicator Expert Skill

Environment

• Python with openalgo, pandas, numpy, plotly, dash, streamlit, numba
• Data sources: OpenAlgo (Indian markets via client.history(), client.quotes(), client.depth()), yfinance (US/Global)
• Real-time: OpenAlgo WebSocket (client.connect(), subscribe_ltp, subscribe_quote, subscribe_depth)
• Indicators: openalgo.ta (ALWAYS — 100+ Numba-optimized indicators)
• Charts: Plotly with template="plotly_dark"
• Dashboards: Plotly Dash with dash-bootstrap-components OR Streamlit with st.plotly_chart()
• Custom indicators: Numba @njit(cache=True, nogil=True) + NumPy
• API keys loaded from single root .env via python-dotenv + find_dotenv() — never hardcode keys
• Scripts go in appropriate directories (charts/, dashboards/, custom_indicators/, scanners/) created on-demand
• Never use icons/emojis in code or logger output

Critical Rules

1. ALWAYS use openalgo.ta for ALL technical indicators. Never reimplement what already exists in the library.
2. Data normalization: Always convert DataFrame index to datetime, sort, and strip timezone after fetching.
3. Signal cleaning: Always use ta.exrem() after generating raw buy/sell signals. Always .fillna(False) before exrem.
4. Plotly dark theme: All charts use template="plotly_dark" with xaxis type="category" for candlesticks.
5. Numba for custom indicators: Use @njit(cache=True, nogil=True) — never fastmath=True (breaks NaN handling).
6. Input flexibility: openalgo.ta accepts numpy arrays, pandas Series, or lists. Output matches input type.
7. WebSocket feeds: Use client.connect(), client.subscribe_ltp() / subscribe_quote() / subscribe_depth() for real-time data.
8. Environment: Load .env from project root via find_dotenv() — never hardcode API keys.
9. Market detection: If symbol looks Indian (SBIN, RELIANCE, NIFTY), use OpenAlgo. If US (AAPL, MSFT), use yfinance.
10. Always explain chart outputs in plain language so traders understand what the indicator shows.

Data Source Priority

OpenAlgo API Methods for Data

Indicator Library Reference

All indicators accessed via from openalgo import ta:

Trend (20)

ta.sma, ta.ema, ta.wma, ta.dema, ta.tema, ta.hma, ta.vwma, ta.alma, ta.kama, ta.zlema, ta.t3, ta.frama, ta.supertrend, ta.ichimoku, ta.chande_kroll_stop, ta.trima, ta.mcginley, ta.vidya, ta.alligator, ta.ma_envelopes

Momentum (9)

ta.rsi, ta.macd, ta.stochastic, ta.cci, ta.williams_r, ta.bop, ta.elder_ray, ta.fisher, ta.crsi

Volatility (16)

ta.atr, ta.bbands, ta.keltner, ta.donchian, ta.chaikin_volatility, ta.natr, ta.rvi, ta.ultimate_oscillator, ta.true_range, ta.massindex, ta.bb_percent, ta.bb_width, ta.chandelier_exit, ta.historical_volatility, ta.ulcer_index, ta.starc

Volume (14)

ta.obv, ta.obv_smoothed, ta.vwap, ta.mfi, ta.adl, ta.cmf, ta.emv, ta.force_index, ta.nvi, ta.pvi, ta.volosc, ta.vroc, ta.kvo, ta.pvt

Oscillators (20+)

ta.cmo, ta.trix, ta.uo_oscillator, ta.awesome_oscillator, ta.accelerator_oscillator, ta.ppo, ta.po, ta.dpo, ta.aroon_oscillator, ta.stoch_rsi, ta.rvi_oscillator, ta.cho, ta.chop, ta.kst, ta.tsi, ta.vortex, ta.gator_oscillator, ta.stc, ta.coppock, ta.roc

Statistical (9)

ta.linreg, ta.lrslope, ta.correlation, ta.beta, ta.variance, ta.tsf, ta.median, ta.mode, ta.median_bands

Hybrid (6+)

ta.adx, ta.dmi, ta.aroon, ta.pivot_points, ta.sar, ta.williams_fractals, ta.rwi

Utilities

ta.crossover, ta.crossunder, ta.cross, ta.highest, ta.lowest, ta.change, ta.roc, ta.stdev, ta.exrem, ta.flip, ta.valuewhen, ta.rising, ta.falling

Modular Rule Files

Detailed reference for each topic is in rules/:

Chart Templates (in rules/assets/)

Quick Template: Standard Indicator Chart Script

import os
from datetime import datetime, timedelta
from pathlib import Path

import numpy as np
import pandas as pd
import plotly.graph_objects as go
from plotly.subplots import make_subplots
from dotenv import find_dotenv, load_dotenv
from openalgo import api, ta

# --- Config ---
script_dir = Path(__file__).resolve().parent
load_dotenv(find_dotenv(), override=False)

SYMBOL = "SBIN"
EXCHANGE = "NSE"
INTERVAL = "D"

# --- Fetch Data ---
client = api(
    api_key=os.getenv("OPENALGO_API_KEY"),
    host=os.getenv("OPENALGO_HOST", "http://127.0.0.1:5000"),
)

end_date = datetime.now().date()
start_date = end_date - timedelta(days=365)

df = client.history(
    symbol=SYMBOL, exchange=EXCHANGE, interval=INTERVAL,
    start_date=start_date.strftime("%Y-%m-%d"),
    end_date=end_date.strftime("%Y-%m-%d"),
)
if "timestamp" in df.columns:
    df["timestamp"] = pd.to_datetime(df["timestamp"])
    df = df.set_index("timestamp")
else:
    df.index = pd.to_datetime(df.index)
df = df.sort_index()
if df.index.tz is not None:
    df.index = df.index.tz_convert(None)

close = df["close"]
high = df["high"]
low = df["low"]
volume = df["volume"]

# --- Compute Indicators ---
ema_20 = ta.ema(close, 20)
rsi_14 = ta.rsi(close, 14)

# --- Chart ---
fig = make_subplots(
    rows=2, cols=1, shared_xaxes=True,
    row_heights=[0.7, 0.3], vertical_spacing=0.03,
    subplot_titles=[f"{SYMBOL} Price + EMA(20)", "RSI(14)"],
)

# Candlestick
x_labels = df.index.strftime("%Y-%m-%d")
fig.add_trace(go.Candlestick(
    x=x_labels, open=df["open"], high=high, low=low, close=close,
    name="Price",
), row=1, col=1)

# EMA overlay
fig.add_trace(go.Scatter(
    x=x_labels, y=ema_20, mode="lines",
    name="EMA(20)", line=dict(color="cyan", width=1.5),
), row=1, col=1)

# RSI subplot
fig.add_trace(go.Scatter(
    x=x_labels, y=rsi_14, mode="lines",
    name="RSI(14)", line=dict(color="yellow", width=1.5),
), row=2, col=1)
fig.add_hline(y=70, line_dash="dash", line_color="red", row=2, col=1)
fig.add_hline(y=30, line_dash="dash", line_color="green", row=2, col=1)

fig.update_layout(
    template="plotly_dark", title=f"{SYMBOL} Technical Analysis",
    xaxis_rangeslider_visible=False, xaxis_type="category",
    xaxis2_type="category", height=700,
)
fig.show()