Horizontally, the effect is variable and most noticeable in stable conditions with more than 20 knots of wind, when standing waves will form. As you probably know, you can recognise the existence of these by lenticular clouds, but you will also see ragged cloud around the peak. These should be avoided at all costs due to their turbulence, especially at the wind speeds that lead to their creation. As well as shockloading, momentary loss of control may occur, not to mention coffee all over the place.
A couple of thoughts for when you’re very high up; how much time it takes to get down if you have a problem, and meeting anyone else at that height on an airway who doesn’t expect you. And oxygen.

Landing Sites
Those on peaks or crests usually present you with more escape routes than any on flanks or valley bottoms so, wherever possible, landings should be made on ground higher than the immediate surroundings, so you can vary the approach according to the wind and have a clear overshoot path. Customers, though, have this annoying habit of wanting to land on the most obscure sites!
Use the windward sides of a slope; leeward sides should only be used in operational necessity, because wind flowing down the slope can increase its apparent angle (you need more lateral cyclic to hold the helicopter in place, and you could run out when you reduce power to lower the downwind skid). Don't forget you will not have the full effects of a ground cushion, if at all. Where conditions allow, go as far to the windward edge as possible, to avoid suddenly finding yourself in dead or reversed airflow (as if on a leeslope) and make overshooting easier. The wind coming over the peak will have increased in speed, due to Venturi effects, so a 15 knot wind can easily become double that, aside from your altimeter misreading.
Finding the wind direction can be interesting if the site is bare and gives you no information, and it doesn’t help that mountain flying tends to take place in high pressure conditions, that is, where the winds are light and variable. We are now talking about local winds, caused by convection, for instance, or katabatic effects, combined with the prevailing wind influenced by the ground, or even a mixture of them all. Even a cloud shadow can increase the speed of a downflowing wind from a cold surface. You could judge its effects on the machine itself, flying round the site with a constant speed and power setting, or a constant altitude, which is otherwise known as a contour crawl, because you use one contour all the way round.
Look at your power settings, whether the air is turbulent, your groundspeed varies, whether you drift or whether the nose yaws into or out from the slope. How much pedal you use to keep straight is a good help – a lot of right pedal means the wind is from the left, for example, and a fair amount of vibration means it is behind you, but it may be a good idea, if you can’t have it at the front, to get the wind off to the side that requires the use of the power pedal (the left, in a 206), in case tail rotor authority becomes a problem. Aft cyclic would indicate a tailwind as well.