A second opinion you can argue with NBS Signal Tool applies one fixed rule set to Nifty, Bank Nifty and Sensex, and shows you every reason it reached its answer.
Trend, momentum and option-chain open interest go in. What comes out is BUY CE, BUY PE or WAIT — with a strike, three targets, a stop, and the working shown in sentences rather than hidden behind a score.
What it is
Three indices, one rule set, no discretion
The same checks run on all three. Nothing is tuned per index after the fact, because a rule that needed different settings on Sensex than on Nifty to look good is a rule that was fitted to the past rather than found in it.
15-minute candles
Trend from stacked EMAs, momentum from MACD and RSI, position from VWAP, and the strike taken from the nearest expiry's chain.
Same checks, wider range
Bigger points per move, so targets and stop come out further apart — they are derived from ATR, not from a fixed number of points.
Same checks again
Run identically, which is the only way the three results can be compared to each other at a glance.
How it works
Five steps, in order, every time
There is no model here and nothing is learned. It is a sequence of checks with a gate at the end, which is why it can always tell you why it said what it said.
Fetch the candles
15-minute bars from your own Zerodha session, with the pre-open auction bar dropped.
Read the trend
The 20- and 50-EMA have to be stacked the same way, and ADX has to clear 20.
Read the momentum
MACD histogram, RSI, and VWAP each vote — and the votes that disagreed are shown too.
Read the option chain
Put-call ratio and where open interest sits, which turns a view on the index into a specific strike.
Issue, or refuse to
Reward-to-risk is computed from ATR first. A setup that does not clear 0.6 is marked not worth it and never becomes a ticket. Most of the day ends here.
The screen
It shows its working
Every image on this site is rendered straight from the app's own canvas, not drawn as a mockup.
What it measured
It has been tested, and the result was not good
This is the part a site like this normally leaves out, so it is on the home page instead. The rule set was run across three years of 15-minute candles — 8,837 signals — and came out roughly break-even before costs and negative after brokerage, the option bid-ask and time decay. Its targets are reached about a third of the time.
This server has logged 1 closed trades of its own (2026-09-09 to 2026-09-09): T1 reached on 0.0%, T2 on 0.0%, T3 on 0.0%, the stop hit on 100.0%. A few dozen trades is a sample, not a verdict — the three-year figures above are the measurement.
It is published so it can be checked, not because it is known to work. The full measurement, and what the test did and did not include →
Boundaries
What it will not do
- ×Place an order. Not now, not with a confirmation, not on a schedule.
- ×Give advice. It is not from a SEBI-registered research analyst or investment adviser.
- ×Hide a losing signal. Closed trades go into the log whichever way they went.
- ×Tune itself to whatever just happened. The rules are fixed in config.
- ✓Show its working — the votes, the dissent, the blocker that stopped a ticket.
- ✓Freeze the levels at entry, so they cannot drift while a trade is open.
- ✓Say WAIT, which is what it says for most of most days.
- ✓Use your own broker session, so your data and your rate limit stay yours.
Getting in
Accounts are created by the owner
There is no public sign-up form. You are given an email and a password, you log in, and then you connect your own Zerodha account — which Zerodha makes a once-a-morning step rather than a once-ever one.