Angel One's SmartAPI is free, including live market data — which makes it the usual starting point for anyone who does not want a monthly API bill while learning. The trade-off is that its login flow works quite differently from Zerodha's, and that difference is where most people get stuck on day one.
This Angel One SmartAPI Python walkthrough covers the whole path with runnable code: authentication, market data, placing an order, and streaming live ticks.
What you need first
- An Angel One trading account.
- An app created at smartapi.angelbroking.com — this gives you an API key.
- Your client code, MPIN, and the TOTP secret you get while enabling two-factor authentication. Store the TOTP secret carefully; it is as sensitive as a password.
- Python 3.8+ and two libraries.
pip install smartapi-python pyotp websocket-clientKeep all four credentials out of your code. Read them from environment variables or a config file your repository ignores.
Logging in: TOTP, not a browser redirect
This is the main difference from Kite Connect. Zerodha sends you to a browser, you log in, and it hands back a request token. SmartAPI instead expects your program to generate the current TOTP code itself and pass it in.
from SmartApi import SmartConnect
import pyotp
api_key = 'your_api_key'
client_id = 'your_client_code'
mpin = 'your_mpin'
totp_secret = 'your_totp_secret'
obj = SmartConnect(api_key=api_key)
totp = pyotp.TOTP(totp_secret).now()
data = obj.generateSession(client_id, mpin, totp)
auth_token = data['data']['jwtToken']
refresh_token = data['data']['refreshToken']
feed_token = obj.getfeedToken()
print('Logged in. Feed token:', feed_token)Practically this means a SmartAPI bot can start itself in the morning without anyone opening a browser — genuinely convenient. It also means your TOTP secret sits on that machine, so treat the server's security as seriously as your account's.
Finding the symbol token
SmartAPI addresses instruments by a numeric symbol token, not by trading symbol. You look these up in the instrument master, which Angel One publishes as a JSON file.
import requests
URL = ('https://margincalculator.angelbroking.com/'
'OpenAPI_File/files/OpenAPIScripMaster.json')
instruments = requests.get(URL).json()
sbin = next(i for i in instruments
if i['symbol'] == 'SBIN-EQ' and i['exch_seg'] == 'NSE')
print(sbin['token'], sbin['symbol'])That file is large. Download it once each morning and cache it — fetching it on every run is slow and unnecessary.
Placing an order
Orders are passed as a dictionary. Start with a single share while you are testing.
order = {
'variety' : 'NORMAL',
'tradingsymbol' : 'SBIN-EQ',
'symboltoken' : sbin['token'],
'transactiontype' : 'BUY',
'exchange' : 'NSE',
'ordertype' : 'MARKET',
'producttype' : 'INTRADAY',
'duration' : 'DAY',
'price' : '0',
'quantity' : '1'
}
order_id = obj.placeOrder(order)
print('Order id:', order_id)
# Kya sach me bhara? Hamesha check karo
for o in obj.orderBook()['data']:
if o['orderid'] == order_id:
print(o['status'], o['averageprice'], o['text'])As with any broker API, a successful call means the order reached Angel One — not that it was executed. Margin shortfalls, circuit limits and frozen quantities all produce rejections. Read the order book back every time.
Streaming live ticks
SmartAPI's WebSocket v2 needs the auth token, api key, client code and feed token from the login step.
from SmartApi.smartWebSocketV2 import SmartWebSocketV2
sws = SmartWebSocketV2(auth_token, api_key, client_id, feed_token)
token_list = [{'exchangeType': 1, 'tokens': [sbin['token']]}] # 1 = NSE
def on_data(wsapp, message):
print(message)
def on_open(wsapp):
sws.subscribe('sub1', 1, token_list) # mode 1 = LTP
sws.on_data = on_data
sws.on_open = on_open
sws.connect()Keep the callback light. Push ticks onto a queue and do your strategy work elsewhere, or you will start dropping data during busy sessions.
Free API, but read this before you switch
SmartAPI costs nothing, including live and historical data, and for someone learning that removes a real barrier. Two things are worth weighing honestly.
First, reliability during live sessions is what you are actually paying for when you pay. Whether a free API holds up under pressure is something you should test with small size before trusting it with a full position.
Second, the flat-fee comparison depends entirely on your trading frequency. If you are placing a handful of orders a month, free wins comfortably. The full cost breakdown across brokers is in what Kite Connect actually costs.
The pragmatic approach that many traders settle on: learn and backtest on a free API, then decide which broker to run live money through once you know how often your strategy actually trades.
Frequently Asked Questions
Is Angel One SmartAPI free?
Yes. SmartAPI has no subscription charge and includes live market data and historical data, which is why it is a common starting point for people learning algo trading in India.
Why does SmartAPI need a TOTP secret?
SmartAPI does not use a browser redirect for login. Your program generates the current two-factor code itself using the TOTP secret and passes it to generateSession. It means the bot can log in unattended, but it also means the secret sits on that machine, so secure it accordingly.
What is a symbol token in SmartAPI?
A numeric identifier for an instrument. SmartAPI takes this token rather than the trading symbol, and you look it up in Angel One's instrument master JSON file. Download that file once a day and cache it rather than fetching it on every run.
Should I use Angel One or Zerodha for algo trading?
It depends on your trading frequency and how much you value execution reliability. Angel One's API is free, which suits learning and low-frequency strategies. A paid API like Kite Connect is easier to justify when live money is deployed daily and downtime is costly.
Related Reading
- Kite Connect Python tutorial - the Zerodha equivalent
- What Kite Connect actually costs
- How to run your trading bot 24x7
- Algo Trading with Python - complete course
Disclaimer: TheFinBaba provides educational content only - this is not investment advice. Trading involves risk of loss.
