Every call arrives with its homework attached.
Entry, position size, stop, targets, and the reason the system took the trade, written down at the moment it was taken. You place the orders yourself, in your own account, on your own platform. Nothing about the decision is hidden from you, including the ones that go wrong.
A signal is five decisions, not one.
Most alerts hand you a ticker and leave you to invent the rest under time pressure. Every ORO SAGE message carries the whole decision, in the same plain format whether it is a win, a loss, or a stop being raised.
- Entry. The price the system acted at, sent only after the broker confirmed the fill. Never on an intention, never on a plan to act.
- Size. What the position is worth as a share of the book, capped by conviction before entry rather than argued about afterwards.
- Stop. A specific price, set before the trade exists, with the worst case stated in rupees so you know what you are risking on the day you take it.
- Targets. Two of them, with the part-exit rule attached, so booking is a rule and not a mood.
- The reason. The conviction score and the checks the candidate passed, written into an append-only journal at the moment of the decision. It cannot be tidied up later to look smarter than it was.
Then the position updates: stop raised, first target hit, exit taken. Each one restates the worst case in plain rupees at that moment, so you always know where the trade can end.
🟢 BUY · ACME LTD (SW): 4 shares @ ₹3,412.00 (₹13,648). Stop set at ₹3,268.50 (−4.2%). Targets: T1 ₹3,742 / T2 ₹4,094.
You know the size, the price paid, and the worst case in rupees before anything else happens: ₹574.
🔼 STOP RAISED · ACME LTD (SW): ₹3,268.50 → ₹3,415. Loss no longer possible: worst case is now +₹12 profit (was ₹574 max loss).
The stop moves at the broker, not in a spreadsheet. From this message on, the trade cannot end in a loss.
🎯 T1 HIT · ACME LTD (SW): sold 2 of 4 @ ₹3,742 → banked +₹660 (+9.7% on those 2). Riding: 2 shares · next target T2 ₹4,094 · stop ₹3,415 (so this trade can no longer end below +₹666).
Part booked, the rest still running, and the new floor stated in rupees. No guessing about what to do next.
Every message carries the strategy bucket (swing, positional and so on), exact quantities and prices, and what the worst case now is in plain rupees. A message fires only after the broker confirms the trade happened, never on an intention. The losses arrive in the same plain format as the wins.
Illustrative example with an illustrative stock. Not a recommendation to buy or sell any security.
The easy ten percent is what to buy.
The other ninety percent is what breaks accounts, and it is the part nobody hands over. You already know this, because you have lived it. Unlike others in the industry, we would rather write it down than pretend it away.
Sizing is where the damage is
The same idea can build an account or wreck it depending on how much of the book it gets. A tip has no size attached, so the size ends up being whatever you felt like that morning. Conviction sets it here, before entry, with a ceiling it cannot exceed.
The stop moves when you are losing
Everyone sets stops. The trouble is that the moment a stop matters is the moment it feels wrong, and a stop that lives in your head can be renegotiated. The system's stop is placed at the broker with the entry, so it is not available for a second opinion.
Exits are decided at the worst time
Booking early, riding too long, averaging down into a falling name: all of it happens in the minutes when judgment is at its weakest. Targets, trails and part-exits are written before the trade opens, so the decision has already been made by someone calmer.
The mechanics, so you can pull them apart.
This is how the system runs its own book, which is where the calls come from. If any of it does not survive your scrutiny, we would rather you found out now than after you paid.
The stop lives at the broker
When the system enters a position, the stop and the target are placed at the broker with the entry, so they are live at the exchange from the first moment. If ORO SAGE goes down, the protection does not go down with it.
Gap-downs, said plainly
A stop is a trigger price, not a promise. If a stock opens below your stop, the order goes live at the open and fills where the market actually is, which can be worse. Nobody engineers that away. What is in our control is that the order was already sitting there before the gap.
It checks the broker, not just itself
The system does not assume an order worked because it asked for it. It reconciles its own records against the broker's own records. If a stop fired at the exchange, the journal records it at the price it actually sold at, not the price we hoped for.
A kill switch that lands in under a minute
One command halts all new buying in sixty seconds or less, whether a human pulls it or the system pulls it on itself. It cannot fail quietly, and a halt stays a halt until a person deliberately clears it. Six named circuit breakers can pull it without asking anyone.
When an order is refused
A refused order is retried a few times, with widening gaps between attempts. If it still will not go through, that name is paused for twenty-four hours instead of being hammered. A cascade of rejections trips its own breaker and stops the session.
One door out
Orders can leave through one controlled, verified route and no other, in line with the algo framework. One door, and one record of everything that walked through it.
It restarts scared
After an interruption the system restarts paused, and it will not trade until its own record of what it holds matches the broker's, position by position. The dangerous state is a machine that resumes trading while it is confused about what it holds.
Fail loud, never fake
A stale or fallback price is never allowed to pass as live. Degraded data is tagged degraded or the run stops. Every value carries its source and its age, all the way through, so a decision can show the receipt for the numbers it acted on.
About 5,700 listed names, 163 data columns each, are screened every trading morning before any of this begins. Figures as of 07-Aug-2026.
Every brake, with its numeric trigger and who resets it, and what happens if ORO SAGE itself goes down. Or read how a single decision moves through the seven roles.
What stays entirely yours.
Signals is a subscription to a system's decisions. It is not access to your account, and it never becomes access to your account.
If you would rather not place the orders at all, the same decisions can run inside your own broker account without you touching anything: that is Autopilot, which opens after registration and broker algo-empanelment.
We publish the bugs too.
The engineering notes in the journal are written the way we would want to read them: what broke, how long it was broken, how it was caught, and what changed. A vendor who only publishes wins is telling you about their marketing, not their system.
The empty-folder bug: our system decided on nothing for 150 cycles
A silent failure that produced clean-looking output while the input was missing. The reason a separate part of the system now does nothing but watch the rest.
TRIGGERED, not TRADED: learning the broker's language before real money
Two broker statuses that look identical and mean opposite things. Found on paper, before a rupee was on the line.
There is a third on the one-route rule for orders and what it took to satisfy it: One door out. The whole set lives in the journal.
Fair objections, answered straight.
If an answer here sounds like marketing, tell us and we will rewrite it.
What happens when it is wrong?
It is wrong often. Every system that trades is, and any vendor implying otherwise is one you should walk away from. What the system controls is the size of wrong. The stop is defined before the position exists, the size is capped before the position exists, and the exit is not a judgment call made while a trade is bleeding. When a stop fires, the message arrives in the same plain format as a win, at the price the broker actually filled, confirmed against the broker's own records. Losses are not quietly dropped from the record. That is the whole point of an append-only journal.
Why would I pay when tips are free on Telegram?
Because a free tip is the easy ten percent. It has no size, no stop, no exit rule, and above all no denominator: you never learn how many calls were made, how many were quietly deleted, or what the losers did. You are the risk manager, the position sizer and the record keeper, for free.
What you would be paying for is the other ninety percent and the record that lets you check it. If that is not worth a fee to you, the free channels are genuinely fine and we would rather you kept your money.
You have no public track record. Why should I go first?
You should not go first on faith, and we are not asking you to. Here is exactly what exists today. Real money has been trading on this system's decisions since April 2026, with the founder placing each order by hand. In August 2026 the system began placing its own orders, on the founder's own capital. Client money is not in it and will not be until registration and a live record worth showing, drawdowns included, are both real.
No performance figures appear anywhere on this site, and none will until that record means something. What you can weigh right now is the rigour behind it: 3,401 automated tests, all green, as of 06-Aug-2026. Not one live order was sent in the system's entire build history until the machine had been proven on paper first.
What if I miss the message?
You skip that one. These are swing and positional calls held over days and weeks, not intraday scalps that die in ninety seconds, so a signal is not a race you can lose by being at lunch. Every message states the worst case in rupees at the time it is sent, so a late reader can see whether the trade still makes sense at the current price rather than chasing a fill. The system's own book carries on either way, and your skipping does not break the sequence.
When can I actually pay for this?
Not today. ORO SAGE is applying for registration under the applicable SEBI framework, and paid services open only once that registration is in place. Autopilot needs broker algo-empanelment on top of it. Until then this page collects expressions of interest and nothing on it is investment advice.
What joining does buy you is order and price: founding members are onboarded first, at the founding price locked to the date you joined.
Signals
The system's decisions, delivered to you. You place the trades in your own broker account.
- Buy, stop-loss and target alerts, with the reason and conviction score attached
- Daily market pulse and watchlist
- Position management updates: when to trail, when to book, when to leave
- You stay in control of every click
Founding pricing, locked to your join date. No profit share, no fee on your assets, no multi-year lock-in.
Request access to SignalsOpens once registration under the applicable SEBI framework is in place.
Autopilot
ORO SAGE runs the whole loop inside your own broker account. It decides, sizes, places and manages. You keep custody.
- Everything in Signals, plus hands-off execution and position management
- Runs through your broker's API. The money never touches us
- Your kill switch, your right to pause, at any time
- The full decision journal, yours to audit whenever you like
Founding pricing, locked to your join date. No profit share, no fee on your assets, no multi-year lock-in.
Minimum deployable capital: ₹2,00,000. Below that, position sizing and the concentration caps cannot do their job, so the system would be running with its hands tied. We would rather tell you that than take the fee.
Request access to AutopilotOpens after registration and broker algo-empanelment are both in place.
Put your name down for Signals. Launch updates and early access, first, with your founding price locked to the date you join.
Launch updates and early access, first. Founding price locked to your join date.