bandl

One Python client for market data — crypto, Indian equities & options.
Same call everywhere. Get a pandas DataFrame or typed bars in three lines.

from bandl import Bandl, Interval

df = Bandl().crypto.get_ohlcv_dataframe("BTC/USDT", Interval.D1)   # no API key needed

One client, one API. Switch markets by changing the symbol — not your code.

• 🟢 Zero-config crypto — Binance & CoinDCX public data, no keys.
• 🇮🇳 Indian equities & indices — RELIANCE, NIFTY 50, BANKNIFTY via Zerodha.
• 📈 Options, incl. expired contracts — NSE/BSE F&O + MCX commodities via Dhan.
• 🐼 pandas or typed models — get_ohlcv_dataframe(...) or get_ohlcv(...) -> list[OHLCV].
• ⏱️ Normalized everywhere — UTC timestamps, Decimal prices, one Interval enum.
• 🤖 Agent-ready — a dedicated AGENTS.md reference for LLM tools.

Install

pip install bandl

Python 3.10+. Dev setup: pip install -e ".[dev]" (CONTRIBUTING.md).

60-second start

No API key required — crypto works out of the box:

from bandl import Bandl, Interval

client = Bandl()

df = client.crypto.get_ohlcv_dataframe("BTC/USDT", Interval.D1)
print(df.tail())
#            timestamp     open     high      low    close       volume
#  2025-01-10 00:00:00  94000.1  95200.0  92800.5  94850.2   12930.4451

Want a window? Pass start / end (UTC). Want raw bars instead of pandas? Call get_ohlcv(...) — same arguments, returns list[OHLCV].

from datetime import datetime, timedelta, timezone

end = datetime.now(timezone.utc)
start = end - timedelta(days=30)
bars = client.crypto.get_ohlcv("ETHUSDT", Interval.H1, start, end)
print(bars[-1].close, bars[-1].source)

One API, every market

Every facet exposes the same two calls — get_ohlcv(...) and get_ohlcv_dataframe(...). Pick a provider with source="...", or rely on each facet’s default.

Recipes

Indian equities & indices (Zerodha)

Add your Kite Connect credentials once; the rest is identical to crypto. Symbol aliases (NIFTY 50 → NIFTY50) are handled for you.

from bandl import Bandl, BandlConfig, Interval, ProviderSettings

client = Bandl(BandlConfig(providers={
    "zerodha": ProviderSettings(api_key="kite_api_key", access_token="daily_token"),
}))

reliance = client.equity.get_ohlcv_dataframe("RELIANCE", Interval.D1, source="zerodha")
nifty    = client.equity.get_ohlcv_dataframe("NIFTY 50", Interval.D1, source="zerodha")

Options & derivatives (Dhan)

client.derivatives serves option OHLCV across NSE/BSE F&O and MCX commodities — with open_interest on every bar. Give it a symbol string (auto-resolved against Dhan’s scrip master) or a structured OptionContract.

from datetime import date, datetime, timezone
from decimal import Decimal
from bandl import Bandl, BandlConfig, Interval, ProviderSettings
from bandl.models.market import OptionContract, OptionType

client = Bandl(BandlConfig(providers={
    "dhan": ProviderSettings(api_key="dhan_client_id", access_token="dhan_jwt"),
}))

# 1) Symbol string — easiest
df = client.derivatives.get_ohlcv_dataframe(
    "GOLDM26JUL145000CE", Interval.M5, source="dhan", exchange="MCX",
)

# 2) Structured contract — explicit & unambiguous
contract = OptionContract(
    underlying="GOLDM", expiry=date(2026, 7, 29),
    strike=Decimal("145000"), option_type=OptionType.CALL, exchange="MCX",
)
bars = client.derivatives.get_ohlcv(contract, Interval.M1, source="dhan")

# What expiries exist for an underlying?
expiries = client.derivatives.list_expiries("GOLDM", source="dhan", exchange="MCX")

Need expired contracts? Most APIs drop them. bandl still fetches their minute candles — pass the native instrument_id once (look it up via Dhan, or the bundled examples/dhan_expired_probe.py):

bars = client.derivatives.get_ohlcv(
    "GOLDM26JUN143500CE", Interval.M1,
    datetime(2026, 6, 26, tzinfo=timezone.utc),
    datetime(2026, 6, 27, tzinfo=timezone.utc),
    source="dhan", exchange="MCX", instrument_id="570800",
)

Indian equities, derivatives & commodities (ICICI Direct Breeze)

ICICI Direct Breeze provides historical OHLCV across equities, indices, and derivatives (including MCX commodities down to 1-second candles), along with Demat portfolio holdings and live order execution.

1. Obtaining API credentials & Session token

• Register/login on the ICICI Direct Breeze API Portal.
• Create an app to receive your App Key / API Key (api_key) and Secret Key (api_secret).
• To generate your daily session token:
  1. Open: https://api.icicidirect.com/apiuser/login?api_key=<YOUR_API_KEY> in your browser.
  2. Log in with your ICICI Direct credentials and TOTP.
  3. After login, your browser redirects to your registered Redirect URL with ?apisession=<session_token>.
  4. Pass <session_token> as access_token and your ICICI User ID as account_id in ProviderSettings.

2. Configuration & Market Data

from datetime import date, datetime, timezone
from decimal import Decimal
from bandl import Bandl, BandlConfig, Interval, ProviderSettings
from bandl.models.market import OptionContract, OptionType

client = Bandl(BandlConfig(providers={
    "breeze": ProviderSettings(
        api_key="your_app_key",
        api_secret="your_secret_key",
        access_token="your_daily_session_token",
        account_id="your_user_id",
    ),
}))

# 1) Equities & Indices (NSE / BSE)
df_rel = client.equity.get_ohlcv_dataframe("RELIANCE", Interval.D1, source="breeze", exchange="NSE")

# 2) 1-Minute / 1-Second Candles for Commodities & Futures (MCX)
df_crude = client.equity.get_ohlcv_dataframe(
    "CRUDEOIL", Interval.M1,
    datetime(2026, 9, 20, tzinfo=timezone.utc),
    datetime(2026, 9, 25, tzinfo=timezone.utc),
    source="breeze", exchange="MCX",
)

# 3) Commodity / Equity Options
contract = OptionContract(
    underlying="CRUDEOIL",
    expiry=date(2026, 10, 16),
    strike=Decimal("5600"),
    option_type=OptionType.CALL,
    exchange="MCX",
)
bars = client.derivatives.get_ohlcv(contract, Interval.M5, source="breeze")

3. Portfolio & Trading

from bandl.models.account.types import OrderSide, OrderType
from bandl.models.trading import OrderRequest, ProductType

# Demat holdings, cash balances, and margin info
holdings = client.portfolio.get_holdings(source="breeze")
balances = client.portfolio.get_balances(source="breeze")
margin   = client.portfolio.get_margin(source="breeze")

# Place a regular order
ack = client.trade.place_order(
    OrderRequest(
        symbol="RELIANCE",
        exchange="NSE",
        side=OrderSide.BUY,
        order_type=OrderType.LIMIT,
        quantity=Decimal(10),
        price=Decimal("2450.00"),
        product=ProductType.DELIVERY,
    ),
    source="breeze",
)
print("Order placed ID:", ack.order_id)

Typed bars instead of pandas

from bandl import OHLCV

bars: list[OHLCV] = client.crypto.get_ohlcv("BTCUSDT", Interval.H1)
bars[-1].close      # Decimal — no float rounding
bars[-1].timestamp  # tz-aware UTC datetime

List tradable symbols

client.list_symbols(source="binance", search="BTC", limit=20)
client.list_symbols(source="zerodha", exchange="NSE",
                    instrument_types=("EQ",), search="RELI", limit=10)

Intervals & timezones

One enum maps to every provider’s native interval. Timestamps come back UTC.

from bandl import Interval
Interval.M1, Interval.M5, Interval.H1, Interval.D1   # 1m / 5m / 1h / 1d

df["timestamp"] = df["timestamp"].dt.tz_convert("Asia/Kolkata")  # → IST for display

Configuration

from bandl import BandlConfig, ProviderSettings

config = BandlConfig(
    providers={
        "zerodha": ProviderSettings(api_key="...", access_token="..."),
        "dhan":    ProviderSettings(api_key="client_id", access_token="jwt"),
    },
    timeout_seconds=30,
    default_crypto_provider="binance",       # client.crypto default
    default_equity_provider="zerodha",       # client.equity default
    default_derivatives_provider="dhan",     # client.derivatives default
)

Binance HTTP 451? Binance blocks some regions/cloud IPs (US, Colab). Use source="coindcx" — same symbols, no key — or set default_crypto_provider="coindcx".

CoinDCX empty DataFrame? Its public feed can lag by months. A start/end entirely after the feed raises DataNotAvailableError with the available span; pick an overlapping window.

More

fills  = client.account.get_fills(start, end, source="coindcx")
pnl    = client.account.get_pnl(start, end, source="zerodha", prefer="auto")
bundle = client.account.export_analysis_bundle(start, end)

from bandl import AssetType

tickers = client.crypto.get_24hr_tickers(source="coindcx", asset_type=AssetType.CRYPTO_PERP)

cp examples/.env.example .env      # add ZERODHA_* / DHAN_* to test authed providers
python examples/main.py
python examples/dhan_options.py
python examples/futures_24hr_leaders.py --source coindcx

For AI agents

AGENTS.md is a purpose-built reference (provider matrix, recipes, errors) for LLM coding tools. It is not shipped in the PyPI wheel — point your agent at the GitHub link:

https://github.com/stockalgo/bandl/blob/master/AGENTS.md

Pin a tag (e.g. .../blob/v0.4.0/AGENTS.md) for a fixed version. See agents/README.md.

Docs & development

• docs/BANDL.md — layout & design notes
• docs/ACCOUNT_HISTORY.md — account facet
• docs/EXECUTION_MIGRATION.md — live execution & recovery guide
• docs/PROVIDER_AUTHOR_GUIDE.md — building execution adapters
• CONTRIBUTING.md — tests, Ruff, pull requests
• SECURITY.md — reporting vulnerabilities

pytest tests/bandl/
ruff check lib/bandl tests/bandl

Roadmap

• Live streams / WebSockets
• More brokers & deeper commodity history
• Richer SymbolInfo and fundamentals

Contributing

PRs welcome — read CONTRIBUTING.md and CODE_OF_CONDUCT.md first.

License

MIT