TradeLens is a personal trading intelligence dashboard — think of it as your own trading command centre. It lives in your web browser and shows you live stock market data, helps you spot trading opportunities (called "recommendations"), lets you track everything you own (your portfolio), and monitors two automated trading bots that buy and sell shares on your behalf.

It connects to real market data feeds, analyses stock patterns, and gives you a single screen to see everything happening with your investments at once — without needing to log into ten different websites.

Open your browser and go to tl.kailorix.com.au. You'll see a login screen asking for a username and password. Use your admin credentials. Once logged in, your session is remembered for the rest of the browser session — you won't need to log in again unless you close the browser or the session expires.

If you forget your password, it is stored in the server's configuration file (settings on the server) and can be reset by updating the server settings.

The Dashboard is your home page — a quick summary of everything. It shows:

• Market summary — how the main markets (ASX 200, S&P 500, etc.) are doing right now
• Top movers — stocks that are moving up or down the most today
• Bot status cards — whether your Longterm and Intraday bots are running
• Recent alerts — any price alerts you've set that have triggered

Think of it like the front page of a newspaper — the most important headlines at a glance.

These are the three stock markets TradeLens covers:

• ASX — Australian Securities Exchange (Australian stocks like CBA, BHP, CSL). Trades 10am–4pm Sydney time.
• US — United States markets (NYSE + NASDAQ), covering global giants like Apple, Tesla, NVIDIA. Trades 9:30am–4pm New York time (which is 11:30pm–6am Sydney time).
• NSE — National Stock Exchange of India (Indian stocks like Reliance, Infosys, TCS). Trades 9:15am–3:30pm Mumbai time.

You can switch between markets using the dropdown at the top of most pages. TradeLens will automatically fetch data for whichever market you select.

Imagine you had a very clever friend who scanned thousands of stocks every day and told you "here are the ones that look interesting right now." That's what AI Recommendations does.

The system scans stocks using technical rules (things like: is the price trending up? is volume higher than usual? does it pass momentum checks?) and ranks them by quality. You enter how much money you want to invest and it tells you: here are the best-looking setups, how many shares to buy, what price to pay, where to put your stop loss, and what your potential profit target is.

Important: These are suggestions, not guaranteed profits. Always review before acting.

Paper trading is like playing Monopoly with fake money, but the prices on the board are real stock prices. You make real buy and sell decisions, your trades are tracked, you see profits and losses — but none of it uses real money.

It's called "paper" trading because in the old days, traders would write their practice trades on a piece of paper and see how they would have done. Today, Alpaca (the broker TradeLens connects to) provides a "paper account" that works exactly like a real trading account — you get $100,000 of pretend money to practice with.

Both bots in TradeLens currently use paper trading accounts. No real money is at risk.

The Reports page shows ALL positions — both open (currently held) and closed (already sold). Click "Apply Filters" with no filters set and you'll see your open positions with a yellow HOLD badge. Completed trades (buys that were later sold) show WIN, LOSS, or BREAKEVEN.

If it still shows nothing, the bots haven't made any trades yet, or the bots are not running and their databases are empty. Check the Longterm Trader and Intraday Trader pages to confirm the bots are active.

Simple rule: Green = good (profit / up), Red = bad (loss / down).

• Green price — the stock went up
• Red price — the stock went down
• Green P&L — you're making money on that position
• Red P&L — you're losing money on that position
• Green dot next to market — market is currently open and trading
• Red dot — market is closed

• Bot pages (Longterm/Intraday Trader) — refresh every 10 seconds while you're on that page
• Stock screener / market summary — background scan every 30 minutes
• Price alerts — checked every time the screener runs
• Reports page — loads fresh from the database each time you click "Apply Filters"
• Portfolio / Watchlist prices — fetched live when you open those pages

Yes! TradeLens is designed to be responsive — it adapts to smaller screens. On phones and tablets, the sidebar collapses into a hamburger menu (the three-lines icon at the top left). Tap it to open and navigate between pages.

The tables on pages like Reports will scroll horizontally on smaller screens — just swipe left/right to see all columns.

Go to the Reports page, apply any filters you want (or leave them blank for everything), then click the "Export CSV" button. This downloads a spreadsheet file that you can open in Excel or Google Sheets containing all your trade data — symbols, dates, prices, P&L, and more.

The Settings page lets you:

• Theme — switch between dark mode and light mode
• Default market — which market (ASX/US/NSE) loads by default
• Brokerage cost — set your per-trade brokerage fee so allocation calculations include it
• Ticker tape — show or hide the scrolling price ticker at the top
• Notifications — manage Telegram alert preferences

The Longterm Trader Bot is an automated stock-buying computer program. Think of it like having a robot financial adviser who watches the stock market every single day, decides when to buy and sell shares, and does it all automatically — without you having to lift a finger.

It's called "Longterm" because it holds positions for days, weeks, or even months (unlike the Intraday Bot which opens and closes trades within a single day). It trades US stocks on the Alpaca paper trading account, starting with $100,000 of virtual money.

It runs once per trading day at 5:01 PM New York time (after the US market closes), analyses market conditions, and decides whether to buy or sell.

It's a multi-step filtering process. Think of it like a casting director hiring actors — many apply, only the best make the cut:

1. Market regime check — is the overall market "healthy" (bull/bullish)? If not, the bot won't buy anything.
2. Momentum screen — is the stock moving strongly upward recently?
3. Correlation guard — is the stock too similar to positions already held? (avoids doubling up on correlated risk)
4. Risk check — does the trade fit within today's risk budget? Is the position size correct?
5. Signal generation — does the stock produce a clean buy signal based on price patterns and volume?

Only stocks that pass every single filter get bought. That's why many days the bot buys nothing — most stocks don't pass all the checks.

A "regime" is the bot's diagnosis of what mood the overall stock market is in right now. It's like a weather forecast — "is it sunny, cloudy, or stormy?"

• STRONG_BULL — Market is surging confidently upward. Best conditions for buying. Bot is aggressive.
• BULL — Market is trending upward. Good conditions. Bot buys normally.
• NEUTRAL — Market is going sideways — neither up nor down clearly. Bot is cautious, may not buy.
• BEAR — Market is trending downward. Danger! Bot will not open new long positions.

The regime is calculated using an HMM (Hidden Markov Model) — a mathematical algorithm that analyses patterns in price, volume, and volatility data to classify market states. Your current regime is STRONG_BULL — the best possible conditions.

HMM stands for Hidden Markov Model. Here's the simple explanation:

Imagine you're trying to guess if it's raining outside without looking out the window. You can't see the weather directly (it's "hidden"), but you can look at clues — is the grass wet? Are people carrying umbrellas? Are cars splashing puddles?

The HMM does the same thing with the stock market. It looks at clues (returns, volatility, volume) and uses probability mathematics to decide: "based on these signals, the market is most likely in a BULL state right now with 80% confidence."

It was trained on years of historical S&P 500 data and learned to recognise patterns that typically precede bull markets vs bear markets. The model runs every day and updates its assessment.

The Longterm Bot runs once per trading day at 5:01 PM New York time (ET), which is:

• 7:01 AM Sydney time (AEST) — next day
• 3:01 AM London time

It waits until 5pm because it uses "daily bar" data — meaning it needs the full day's trading data (open, high, low, close, volume) to have finished before it analyses anything. If it ran at 2pm, the day's data would be incomplete.

On weekends and US public holidays, the bot detects that the market is closed (using Alpaca's calendar API) and skips that day entirely, sending you a Telegram notification saying "Market closed today, skipping."

A stop loss is like an automatic safety net. Imagine you buy a stock at $100 and you decide: "If it falls to $90, I'll cut my losses and sell — I don't want to lose more than $10 per share." The stop loss is your $90 line in the sand.

If the stock drops to $90, the bot automatically sells it, limiting your loss to $10 per share. Without a stop loss, you might watch a stock fall from $100 all the way to $40 and lose $60 per share.

The Longterm Bot calculates stop losses using ATR (Average True Range) — a measure of how much a stock typically moves per day. A more volatile stock gets a wider stop so it doesn't get knocked out by normal daily swings.

A regular stop loss stays fixed at one price. A trailing stop is smarter — it moves up as the stock price rises, locking in more and more profit.

Example: You buy a stock at $100, set a trailing stop 5% below the current price ($95). Stock rises to $120 — your stop now moves up to $114 (5% below $120). Stock keeps rising to $150 — your stop moves to $142.50. If the stock then drops to $142.50, you automatically sell and lock in that profit.

This way, winners keep running and you capture as much upside as possible, while still being protected if the stock reverses.

A circuit breaker is an emergency stop button. Just like your home's electrical circuit breaker trips when there's a power surge (to prevent a fire), the trading bot's circuit breaker stops all trading when things are going badly — to prevent a financial disaster.

For the Longterm Bot, if the portfolio falls by more than a set percentage in a short period, the circuit breaker trips: no new positions are opened until conditions improve or a reset is triggered.

You can see the circuit breaker status on the Longterm Trader page. If it's tripped, it will show a warning and the reason why.

P&L stands for Profit and Loss. Unrealised means "on paper — you haven't actually sold yet."

Example: You bought TSLA at $366. It's now worth $376. Your unrealised P&L is +$10 per share × 10 shares = +$100. But you haven't sold yet, so that profit isn't "in your pocket" — it could go up or down before you sell. Once you sell, it becomes realised P&L — actual money made or lost.

The Reports page now shows unrealised P&L for your open HOLD positions (in yellow) so you always know how your open trades are performing.

A "bar" is one unit of trading data — for the Longterm Bot, one bar = one full trading day's worth of data (open price, high, low, close, volume).

When bar_count = 0, it means the bot ran its daily check but found zero qualifying new signals. This is completely normal — most days, no new stock passes all the filters. It doesn't mean anything is wrong. The bot ran fine; it just had nothing good enough to buy today.

Bar_count would be higher (e.g., 1 or 2) when a bar successfully triggers a signal that leads to a new position being opened.

This is the bot's margin class — a tier that describes the size of the portfolio. "Advanced $100K+" means the portfolio started with $100,000 and is currently in the advanced tier.

Different tiers can have different risk parameters — a larger portfolio might take bigger positions. Your current portfolio value is approximately $100,254 (up from the $100,000 starting amount, so you're slightly in profit from the unrealised gains on MSFT, NVDA, and TSLA).

Alpaca is a US stockbroker that has an API — essentially, a way to give it instructions via computer code instead of clicking buttons on a website. The bot connects to Alpaca's API using your API keys (like a secret password for the code to authenticate with Alpaca), and when it wants to buy or sell, it sends an order to Alpaca.

Alpaca executes the order in the market and confirms back. The bot receives the confirmation and records the trade in its database.

Currently both the Longterm and Intraday bots use separate paper trading accounts — meaning they're connected to Alpaca but using virtual money, not real money.

On April 27, 2026, these three stocks passed all the bot's entry filters simultaneously:

• The overall market was in STRONG_BULL regime — conditions were ideal for buying
• Each stock showed strong upward momentum
• Each passed correlation checks (they're in different enough sectors)
• Each had a valid signal (price and volume patterns confirmed)
• Risk sizing was acceptable for each position

The bot doesn't have opinions or preferences about companies — it purely follows the mathematical signals. If Netflix and Amazon passed the filters instead, it would have bought those.

Both bots send you messages through Telegram — the messaging app. When something important happens (a new trade opened, a trade closed, a circuit breaker triggered, the market is closed), the bot sends you a Telegram message automatically.

You configured your Telegram bot token and chat ID in the system settings. The messages include emojis and key trade details like: which stock, how many shares, what price, what the P&L was, and what triggered the action.

For the Intraday Bot, you'll get messages like: 📈 [IDT Paper Alpaca] LONG TSLA @ $376 | Stop $370 | ORB strategy

The Intraday Trader Bot is a computer program that buys and sells stocks within a single trading day. It opens positions in the morning and closes them before the market closes at 4pm New York time — it never holds a stock overnight.

Think of it like a day trader who wakes up at 9:30am, makes quick decisions based on how stocks are moving, trades through the day, and has cashed out by 4pm. The bot does this completely automatically — no human input required once it's running.

It uses three different trading strategies (ORB, VWAP, and Momentum) and picks the best stocks each morning using a pre-market scanner.

ORB = Opening Range Breakout. Here's the simple version:

Every stock has a "opening range" — the highest and lowest prices it trades between 9:30am and 9:45am (the first 15 minutes of market open). Think of this as the stock establishing its morning "comfort zone."

If the stock then breaks above the top of that range with strong volume, it often means big buyers have arrived and the stock is about to run higher. The bot buys at that breakout moment.

If the stock breaks below the bottom of that range, it often means the stock is about to fall. The bot shorts (bets against) it.

The ORB strategy only runs within the first 90 minutes of market open — after that, the morning momentum has faded.

VWAP = Volume-Weighted Average Price. This is the "average price" for the day, but adjusted for how much trading happened at each price level. It's the fairest measure of "what price is the stock really at today?"

The strategy works like this: when a stock is trending upward and its price dips back toward the VWAP line (within 0.2%) on a quiet pullback, it's often a great buying opportunity — like catching a wave that just pulled back slightly before surging forward again.

The bot enters when: (1) stock has been making higher prices for 3 bars in a row, (2) price has pulled back close to VWAP, and (3) volume is elevated. It sets a stop 1.5× ATR below and targets 3× ATR above.

Momentum trading is the idea that "stocks in motion tend to stay in motion" — kind of like a ball rolling downhill. If a stock is moving strongly in one direction, there's a good chance it keeps going.

The bot uses two indicators to confirm momentum:

• RSI (Relative Strength Index) — measures how "overbought" or "oversold" a stock is on a scale of 0–100. When RSI crosses 50 from below, it means momentum is flipping from negative to positive — a buy signal.
• ADX (Average Directional Index) — measures how strong the trend is (regardless of direction). ADX above 25 means the trend is strong enough to trade. This prevents trading in choppy sideways markets.

Both conditions must be true simultaneously for the bot to enter: RSI crossing 50 AND ADX above 25 = confirmed momentum buy.

The Dynamic Scanner runs every morning at 9:15 AM ET (15 minutes before market open) and selects which stocks to watch that day. Here's how:

1. Fetches live data for ~85 stocks across 7 sectors (Tech, Financials, Healthcare, Energy, Consumer, Industrials, Materials) plus ETFs
2. For each stock, calculates:
   • Gap % — how much did the stock gap up/down from yesterday's close? (Big gaps = big interest)
   • RVOL — how does pre-market volume compare to the usual amount? (High RVOL = lots of interest)
   • News flag — was there a news article about this stock in the last 12 hours?
3. Scores each stock: 40% gap weight + 35% RVOL weight + 25% news weight
4. Takes the top 20 stocks (max 3 per sector) + always adds SPY and QQQ as anchors

This becomes the day's watchlist — only these stocks will be scanned for trade signals during the day.

Volume is how many shares of a stock were traded in a given period. Relative Volume (RVOL) compares today's volume to the "normal" volume for that time of day.

Example: A stock normally trades 100,000 shares in the first hour. Today it's already traded 400,000 shares by 9:30am. Its RVOL = 4.0 (4× the normal amount). That means something is happening — big news, earnings, a major investor buying/selling — and there's unusually high interest in this stock.

High RVOL stocks are preferred because they have better liquidity (easier to buy and sell quickly) and are more likely to make a strong directional move during the day.

There are two independent safety thresholds:

Strategy-level suspension (each strategy tracked separately):

• 5 consecutive losing trades by the same strategy (ORB, VWAP, or Momentum), OR
• Win rate below 45% over the last 20 trades for that strategy

Either condition immediately suspends that strategy. The other two strategies keep trading normally.

Bot-level daily halt (all strategies stopped):

• 3 consecutive losing trades across ALL strategies combined — stops all new entries for the rest of the day
• OR daily P&L hits -1.5% of starting capital — same effect

Both daily limits reset at midnight. Suspended strategies require manual review before being re-enabled.

There is no warning phase — the suspension is immediate when a threshold is crossed. You will be notified via:

• Telegram alerts sent automatically:
  • ⚠️ [INTRA] Strategy FLAGGED: ORB — 5 consecutive losses
  • ⛔ [IDT Paper Alpaca] CONSECUTIVE LOSS HALT — 3 back-to-back losses
  • 🚨 [IDT Paper Alpaca] DAILY CIRCUIT BREAKER — daily -1.5% limit hit
• TradeLens Dashboard — the Strategy Health panel on the Intraday Trader page shows each strategy's status badge (ACTIVE / SUSPENDED) with win rate and consecutive losses, updated every 10 seconds.

There is no warning-before-suspension. The 5th consecutive loss triggers suspension immediately on that trade's close.

The intraday bot has several safety mechanisms that stop trading when things go wrong:

• Daily loss limit — if total daily P&L reaches -1.5% of starting capital, all new entries are blocked for the rest of the day
• Consecutive loss halt — 3 losing trades in a row (across all strategies) halts new entries for the day
• Max simultaneous positions — never more than 3 open positions at the same time
• No trade after 2:30 PM ET — the bot stops opening new positions in the last 90 minutes of market (to ensure there's time to exit before close)
• No trade in first 5 minutes — the first 5 minutes (9:30–9:35 AM ET) are wild and unpredictable; the bot waits for prices to stabilise

An R-multiple measures how much you made (or lost) relative to how much you risked. "R" stands for Risk.

Example: You buy a stock at $100 with a stop loss at $95 — so your risk is $5 (this is "1R"). If the stock reaches $115 and you sell, you made $15 profit. $15 ÷ $5 = +3R — you made 3 times what you risked. If the stock hit your stop at $95, you lost $5 = -1R.

• +3R or better — excellent trade. Even if you win only 1 in 3, you're profitable.
• +1R to +3R — good trade. Profitable if win rate is above 50%.
• 0R — broke even.
• -1R — lost exactly what you risked (normal stop out).
• Below -1R — slippage or gap caused a worse-than-planned exit.

MAE = Maximum Adverse Excursion — the worst point the trade went against you before you exited. This is the deepest red it got while you were still in.

MFE = Maximum Favourable Excursion — the best point the trade went in your favour. This is the peak green it reached before you exited.

Why these matter: If your MFE is much bigger than your final P&L, it means you're exiting too early and leaving money on the table. If your MAE is often close to your final loss, your stops are well-placed. These metrics help identify whether the strategy's exits need improvement.

Example: A trade closes at +$20 profit, but MFE was $80 — it reached $80 green and you only captured $20. That might mean exits are too conservative.

ATR measures how much a stock typically moves (in dollar terms) during a given period. It tells you: "on an average day, this stock moves this many dollars from its low to its high."

Example: TSLA has an ATR of $15. That means in a normal day, Tesla's price swings $15 from its lowest point to its highest. So if you set a stop loss $5 below entry on TSLA, you'll probably get stopped out by normal daily movement — not because anything went wrong. You'd want a stop of at least 1× ATR ($15) away to avoid random noise.

The bots use ATR to set smart stop losses and take-profit targets that are proportional to each stock's own typical movement — a volatile stock gets a wider stop, a calm stock gets a tighter one.

The Correlation Guard prevents the bot from putting all its eggs in one basket.

Imagine you own stocks in NVIDIA, AMD, and Intel — all chip makers. If there's bad news about the semiconductor industry, ALL THREE will fall together at the same time. You'd lose money on three positions at once, even though you thought you were "diversified."

The Correlation Guard measures how similarly two stocks move historically. If the new stock you want to buy is very similar to one you already hold (correlation above a threshold), the bot refuses to open that new position. This keeps your positions truly independent and reduces the risk of multiple simultaneous losses from a single event.

Just like the Longterm Bot checks if the overall market is in a healthy regime before trading, the Intraday Bot does the same. It measures intraday market conditions — is the S&P 500 (via SPY) trending up today, or is it chaotic and choppy?

On days when the overall market direction is unclear or strongly bearish, the bot reduces activity. On clearly trending days (up or down), the momentum and ORB strategies work much better — the regime filter ensures the bot is only aggressive when conditions are favourable.

SPY and QQQ are always included in the watchlist as "anchors" — they're used as market health gauges throughout the day, not just for trading.

Sentiment analysis reads news headlines about stocks and tries to figure out: is the news positive or negative for this stock?

The scanner checks Alpaca's news feed for articles published in the last 12 hours. If a stock has recent news, it gets a "news flag" which boosts its scanner score — because news-driven stocks tend to move more aggressively during the day, creating better trading opportunities.

The scanner doesn't analyse whether the news is bullish or bearish in detail — it just uses the presence of news as a signal of heightened interest and potential volatility. Combined with the gap and RVOL signals, news-flagged stocks often offer the best setups.

There are actually two levels of "paper" here — which can be confusing:

• Internal simulation mode (INTRA_TRADING_MODE=paper) — the bot calculates trades and records them in its database but never contacts Alpaca at all. All P&L is purely simulated based on real price data. No orders go anywhere.
• Alpaca paper mode (INTRA_TRADING_MODE=live + paper Alpaca keys) — the bot sends real orders to Alpaca's paper trading API. Alpaca executes the orders against real market prices and records them in your Alpaca paper account. No real money involved, but orders flow through a real system.

The intraday bot is currently set to Alpaca paper mode — orders are submitted to your dedicated Alpaca intraday paper account and tracked both in TradeLens and in Alpaca itself.

The bot uses risk-based position sizing. Here's the logic:

1. Define maximum risk per trade: e.g., 1% of total capital = $1,000 (on a $100k account)
2. Calculate stop loss distance: e.g., entry at $100, stop at $95 = $5 per share risk
3. Shares to buy = Max risk ÷ Risk per share = $1,000 ÷ $5 = 200 shares

This ensures every single trade risks the same fixed percentage of the account — regardless of which stock, what price, or how volatile it is. A $5 stock and a $500 stock both risk the same dollar amount per trade.

All times are US Eastern Time (ET):

• 9:15 AM ET — Pre-market scanner runs: picks today's watchlist
• 9:30 AM ET — US market opens
• 9:35 AM ET — Bot begins scanning for signals (first 5 minutes skipped)
• 9:30–11:00 AM ET — ORB strategy active (opening range breakouts)
• 9:35 AM–2:30 PM ET — VWAP and Momentum strategies active
• 2:30 PM ET — No new positions opened after this time
• 3:45–4:00 PM ET — All open positions automatically closed
• 4:00 PM ET — US market closes; bot stops for the day

In AEST (Sydney time): 9:30 AM ET ≈ 11:30 PM or midnight the night before (depending on daylight saving).

The first 5 minutes of US market open (9:30–9:35 AM ET) are extremely chaotic. Imagine the stock market is like a busy train station at rush hour — hundreds of orders backed up overnight all try to fill at once. Prices can swing wildly by large amounts in seconds.

Even experienced professional traders avoid the first 5 minutes because:

• Spreads are very wide (big gap between buy and sell price)
• Prices are jumpy and not representative of the actual trend
• Order fills are unpredictable

After 9:35 AM, the market calms down and prices start reflecting real supply and demand. The bot waits for this stability before trusting any signals.

Typically 2–8 trades per day, depending on market conditions. Some days fewer, some days more.

Limiting factors:

• Maximum 3 simultaneous open positions
• Quality filters — most signals are rejected if conditions aren't ideal
• Circuit breakers — if 3 trades lose in a row, trading stops for the day
• No trading after 2:30 PM ET

The bot trades quality, not quantity. It's better to make 2 great trades than 20 mediocre ones.