This guide is written for customers who want one single page that explains what to do, in what order, and why. Follow it from top to bottom. If you do not want to use real money yet, stop at the paper trading section and practice there first.
Install the app, activate the license, and confirm the main UI opens without errors.
Choose how signals will enter the bot: AmiBroker, MT4 or MT5, WebSocket, or manual testing.
Set broker connection, money management, segment windows, and any square-off rules before testing.
Enable paper trading, send test signals, inspect costs, and validate the strategy behavior.
Use history and analytics to verify win rate, costs, behavior during market hours, and consistency.
Turn on live trading only after paper mode results are acceptable and your safeguards are confirmed.
Full setup, cost model, viability tiers, metrics, and paper-to-live decision framework.
Real execution flow, broker safeguards, live controls, and first-week operating checklist.
How stale-feed detection and protection guardrails work and how to verify them.
Start the bot, complete license activation if required, and confirm the main dashboard loads normally.
If activation fails, fix that first. Nothing else matters until the app is licensed and stable.
Decide how the bot will receive entries and exits.
| Signal source | Best use | What to do |
|---|---|---|
| AmiBroker | Strategy automation from AFL | Configure the gateway or plugin, then send a small paper signal to confirm symbol mapping. |
| MT4 or MT5 | MetaTrader-based strategies | Set up the WebSocket or DLL integration and verify the bot receives data from your terminal. |
| Manual sender | Testing and debugging | Use manual signals to validate the bot flow before wiring a live external source. |
If you are still testing, keep your focus on paper trading first. If you plan to go live later, prepare the broker integration now so you do not scramble later.
Supported broker architecture in the project includes major Indian broker workflows such as Shoonya or Finvasia, Zerodha, Fyers, Upstox, and SASOnline.
This is where you define sizing and limits: capital allocation, risk per trade, daily loss caps, weekly caps, and any execution constraints.
The system should be treated as risk-first, not signal-first. A strong signal with bad sizing is still bad trading.
Make sure the correct session or segment rules are configured. Different market segments can have different entry windows and square-off times.
This prevents new trades from being accepted outside valid trading hours and keeps end-of-day behavior predictable.
Once configuration is complete, switch to paper mode and run multiple controlled signals. Confirm entries, exits, rejections, stop-loss behavior, and cost calculations.
Do not judge the system from a single signal. Review a group of paper trades and check whether results, costs, and trade quality make sense over time.
Once paper testing proves the setup is correct, activate live mode carefully and begin with conservative size. Keep the same risk controls.
Paper trading is where you prove the signal flow, cost assumptions, and bot decisions before real execution. Use it to validate the entire chain.
The Trade Calculator is not a toy. It uses the same paper-trading cost model and viability logic so you can test a trade before sending it.
| Scenario | What you should do |
|---|---|
| Trade is rejected | Check the viability score, required edge, and whether your target is too close. |
| Net profit is too low | Increase target quality, reduce weak setups, or review quantity and charges. |
| Auto RR target is higher than your manual target | Your manual target may be too small for the system risk model. |
| Minimum viable TP is much higher than expected | The trade may be too expensive relative to edge, especially on low-range moves. |
| Problem | Likely reason | What to check first |
|---|---|---|
| No trades are appearing | Signal source is not connected or symbols are mismatched | Check the source integration, symbol naming, and whether the bot is actually receiving signals. |
| Signals are rejected | Out-of-window time, weak viability, or blocked risk settings | Review trading windows, score or tier, required edge, and daily loss protections. |
| Quantity looks wrong | Money management settings are misconfigured | Recheck capital allocation, per-trade risk, lot sizing, and broker quantity rules. |
| Profit is much lower than expected | Costs, spread, slippage, or unrealistic target assumptions | Use the Trade Calculator to inspect cost breakdown and minimum viable TP. |
| Paper mode works but live does not | Broker auth, order permissions, or execution mapping issue | Validate broker session, credentials, product type, and symbol or instrument mapping. |
Install the bot. Connect the signal source. Configure risk. Validate on paper. Review the results. Then go live.
Most user mistakes happen because they reverse that order. This guide is designed to prevent that.