BUILD TRADING LOGIC like a puzzle

No. Quberas does not sell "guaranteed returns." The platform gives traders a different kind of tool: a way to formalize a trading idea, test it on historical data, eliminate impulsive decision-making, and run a strategy according to predefined rules. Results depend on market conditions, strategy logic, fees, liquidity, volatility, and risk management. Quberas helps you make decisions more systematically, but it does not eliminate market risk.

No. We do not accept deposits, hold your funds, or have the ability to move assets to external or internal
addresses. Your funds remain in your exchange account, which Quberas connects to through the exchange
API. The API only allows the platform to retrieve the data it needs and send trading commands; it does not
give us access to your funds. In addition, we do not allow API keys with withdrawal permissions to be used
or stored in the system.

Quberas only requires read and trade permissions. Read access allows the platform to see the data needed
for the strategy to work: balances, open positions, order history, and market data. Trade access allows the
bot to place and cancel orders according to its logic. Withdrawal permission must not be enabled - any key with withdrawal rights will be rejected when you try to save it.

Security is layered. The core principle is that an API key must not have withdrawal permissions. We also recommend creating a dedicated API key for Quberas, not reusing the same key across services, enabling two-factor authentication on the exchange, and restricting the API key by IP address if the exchange supports it. The keys themselves are stored on our side in a protected secrets vault (HashiCorp Vault).

If the connection to the exchange is temporarily unavailable, Quberas cannot send new commands, update orders, or close the position until connectivity is restored. Orders that were already placed on the exchange continue to exist according to that exchange's rules. That is why higher-risk strategies should use protective logic in advance: position-size limits, stop conditions, limits on the number of trades, and outage alerts. These events are not fully controlled by Quberas, so every strategy should account for external infrastructure risk.

Quberas alerts you about events that matter for strategy control. Key notifications include: bot start and stop, position opened, position closed, order placed, order filled, order canceled, exchange error, connection loss, risk limit exceeded, stop condition triggered, and trading scenario completed. Notifications are designed to show what the bot did, why it happened, and whether your attention is required.

Yes. The core strategy logic should be the same. A backtest checks a trading strategy on historical market data. If the strategy includes entry, exit, averaging, stop, take-profit, or risk-limit conditions, the backtest applies the same logic to past market data. But a backtest is a simulation. Live trading also depends on order execution on the exchange: liquidity, spread, latency, partial fills, fees, and market behavior at the time of the trade.

Because historical simulation and the live market are not the same thing. In a backtest, a strategy is tested
on data that is already known. In live trading, price moves in real time, an order may not be filled in full, may
be executed at a different price, or may occur during a period of low liquidity. Results are also affected by
fees, spread, slippage, market depth, API latency, exchange response time, order type, and the accuracy of
historical data. The more active the market and the lower the liquidity, the larger this difference can be.

OHLCV is candlestick data: open, high, low, close, and trading volume for the selected period. This data is
suitable for most strategies that operate on candles – for example, 1 minute, 5 minutes, 1 hour, or 1 day.

L1 is the best available bid and ask. Bid is the price buyers are willing to pay. Ask is the price sellers are
willing to accept. L1 is useful when it is important to account for the current spread and model a more
realistic entry or exit price.

L2 is market depth: price levels and order sizes on both the bid and ask sides. L2 is used for more accurate
execution modeling, especially for strategies with limit orders, scalping, large positions, or sensitivity
to liquidity.

Yes. This type of logic is important for proper risk management. One bot can work with multiple trading
pairs if the strategy is designed for multi-pair trading. At the same time, it is important to limit the number
of simultaneous positions, the maximum size of a single position, the total amount of capital in use, and
the number of entries per pair. Risk management is a system of constraints that prevents a strategy from
going beyond an acceptable level of risk. For example, if a signal appears on 20 pairs at the same time,
the bot will not open 20 positions if you allowed only 3 in advance.

The strategy marketplace is a Quberas section where authors can publish their trading strategies, and other users can connect them for review, testing, or use.The strategy author is responsible for its logic: entry and exit conditions, averaging, take-profit, stop-loss, and other rules.
Quberas is responsible for the platform layer: publishing, backtesting, access management, payments, and technical execution.Important: a published strategy is not a guarantee of profit. Even if it performed well on historical data, live trading results may differ because of market risk, fees, liquidity, slippage, and future price behavior.Payment may be structured as a one-time purchase, a strategy subscription, or another format available inside the platform.
The author receives compensation when their strategy is used, while Quberas retains a platform commission. The commission may depend on the strategy type, author plan, and publishing terms.