Has anyone found wrong adjusted prices in NSE historical data for backtesting?

Algos, strategies, code
1

Fellow algo traders — quick question from someone building systematic strategies on
Nifty 50 stocks.

I keep running into a recurring problem: adjusted prices are wrong after corporate
action events — bonus issues, splits,rights issues.

Specific example: yfinance mishandled the HDFCBANK demerger. My backtest was showing
completely wrong returns for 2 years of HDFCBANK data before I caught it.

Three questions for the community:

  1. Which data source do you currently use for NSE historical adjusted OHLCV data?
    (yfinance / EODHD / Kite API / something else?)

  2. Have you ever found wrong adjusted prices that broke a backtest?
    Which stock, which corporate action?

  3. If a clean, correct NSE adjusted data API existed — guaranteed accurate after every
    corporate action — would you pay for it? What would make you switch from your
    current solution?

Genuinely trying to understand how widespread this problem is
before investing time building a solution.

Regards
Primexa

2

i tried multiple times for proper corporate action adjustment, but i failed so many times so i just give up, and consider the change in backtest and live deployment
and corporate action isnt that it came very often, only splits cause the major issue

i havnt found any source which gives adjusted corporate actions, if u find one please share with us