do not use a bot. learn the dark art of trading instead.
why? because
1- bot does not mean AI, and there is no AI that can outtrade a human brain yet. so your run-of-the-mill MT4/5 bots/EAs are not gonna be able to handle those market situations where you would be better. (although James Simmons, ex-NSA cryptographer, mathematician extraordinaire, a genius basically, who created the Renaissance Technologies hedge fund and made billions in the process, might disagree with me. i suspect those guys must have created some amazing black box system siphoning all those billions out of the markets. and you don't see mr simmons trying to peddle some rinky dinky MT4 EA on the web, now do you? exactly.)
2- real bots/EAs are those used by the quant/HFT shops and hedge funds. they have some of the brightest PhDs in math and physics paid top dollar to come up with the best automated systems imaginable... and guess what? they still manage to lose money and can't handle a market meltdown. so if the pros can't pull it off, what makes you think that the EA you'll buy online is gonna do better? if that were the case, that EA would not be for sale anymore and the HFT/hedge fund guys would have bought it and taken it off market. (now that being said, sure, an EA can make you 3 or 5 pips over and over. but can't you make 5 pips yourself? the thing is, most people want a 24/6 5 pip machine. would be nice. but those EAs running on MT4, and everything running on the MT4 platform is simplistic due to the compute limitations of the platform, can't handle market shocks very well and either at best stop trading at no or little loss, or at worst, simply blow up with extreme prejudice.) finally, those guys selling EAs are not in the charity business. so learn to trade first. it's not that difficult and it's fun.
3- damn, i forgot what the 3rd point was. ah yes! implied volatiltiy in the currency market is quite low, meaning profit opportunities are less numerous when compared to trading more volatile instruments such as equities and some commodities. and since we know that most of the time, the FX market is ranging, and a few times trending (depends on what time frame we are looking at, of course), then your bot would need to be really good at trading ranging market conditions. here is a system i'll give you for free to do just so: overlay 3 bollinger bands, standard deviation 1, 2, and 3 and make sure you display the mid-line as well. use an oscillator as well at the bottom of your screen to identify overbought/oversold conditions (i prefer the StochRSI as opposed to the more classic Stoch). and also display the MACD with parameter 8, 17, 9 (DiNapoli's parameters). select M1 as your timeframe. now, make sure the market is not trending because if it is, you'll be dead trying to do this. let's assume that the market is indeed in consolidation mode. choose a currency pair that will oscillate up and down 40 to 70 pips over a couple of hours before another major breakout. now, on M1 you should clearly see that from flat, candles increase in size and eventually hit the 3 standard deviation band (an extreme)--remember that on M1 this unfolds quite quickly, so you don't have time to doze off. immediately check that your oscillator is OB or OS. (if the MACD shows a divergence, great, it's more confirmation. if not, forget about the MACD.) also, check that when the price hits the extreme deviation, preferably it would be close to the 21 or 100 or 200 SMA and/or to a significant fibonacci retrace level, or projection if you identified an AB=CD structure or any other harmonic structure (if no AB=CD, that's OK. if not fib retrace level, OK too. i just use the fibs to give me a higher sense of probability since it seems so many traders are acting based on fibonacci levels. also, many traders reverse trade a move when price hits the 21/100/200 SMA. if price slices through the SMAs, stay with it by keeping an eye on price's relative position to the 3rd standard deviation band and candle shape (more later)) so when all these criteria are met, immediately and without 2nd guessing and afterthought, hit the trigger and sell/buy. do not over leverage because 10 pips in the wrong direction quickly turn into more than your max loss limit of 2.5% of your trading account.
where to exit the trade? as long as candle is flirting with 3rd standard dev or 'walks the band' between 3rd and 2nd standard dev, do not close the trade BUT watch closely to see if you suddenly get a long wick (or spike) with price immediately coming back to the candle body, perhaps this happens when the minute ends and a new candle appears and this new candle is the opposite in direction than the previous one, meaning price is reversing. since the market is not trending but ranging, you want to see these spikes into the extreme band because when they happen the probability of market continuing to stretch and deform the bollinger band in the 3rd standard deviation is extremely low, meaning that the market will reverse, probably going back to the mean, that is, the middle bollinger band, and usually if the oscillator is also coming off an extreme, then to the other side of the bollinger band volatility channel, and the oscillator will go from one extreme (OB for ex) to the other extreme (OS, and vice versa, hence it oscillates, up and down). all you need do is buy/sell at those extremes for 5, 10, 20, 30, or even 50 pips sometimes. 1 mini-lot won't make a lot of money. but if you can use 10 standard lots, that's not a bad day's work as 10 lots mean 1 pip = $100. of course, if using 10 lots, say on a major pair, your margin will be around $12-13,000 or so (100:1 leverage), so to risk no more than 2.5% of your account you would need a total equity in the account of $50,000 if you are risking 12.5 pips = $1,250 (2.5% of $50,000 = $1,250).
and like i said, no time to think. you gotta be quick and not prone to hesitation. otherwise forget it. and also, if the market start to trend all of a sudden, then you may lose your 12.5 pips but if you have been successfully banking more than that during the day, who cares. you still made a sizeable profit.
also, to set all this up, consider the time when you are trading. right at the beginning of the European and North American session, before a trend will emerge (if it will) price discovery will occur meaning price will go up and down within a range. but could be dangerous of course if we have a suddenly violent move. otherwise, outside of those hours, market will be ranging with very low probability of a trend, which is safer to scalp.
hope this helps and godspeed. (use tradingview.com to chart the market. awesome app. best execution at LMAX.)