Comments
If someone knew personally a couple of good, switched on people running a small NGO on the ground (potentially EA adjacent people), my insinct would be to go for that as it will likely be more efficient and well targeted than later scale efforts
From my experience in the aid world, MSF do an amazing job of counterfactually providing medical care where noone else does or can. In difficult conflict situations I suspect their cost effectiveness may sometimes be surprisingly high as there are likely to be many people who would have died if they weren't there.
My instincts and prior research has pushed me away from most BINGOs (big international NGOs) like world vision, care international, save the children etc as they are inefficient generalists, without special expertise and are more inclined to follow the whims of donor money with their work rather than double down on their competitive advantage or high impact work on the margins. I don't see why this would be different in a warzone.
I really like this frame of "given these constraints, how can we do the most good?", and I think EA doesn't do enough of that.
Giving the standard EA answer: Usually, people in crisis situations are less neglected than the global poor, that also lack access to clean water and healthcare. Right now, millions of people are thinking about how they can help Palestinians. So you should still prefer donating to e.g. GiveWell recommendations.
However, that misses the point that there are cheap effectiveness improvements you can unlock by slightly improving your friends' decision making and having them select a better Palestinian org.