I'm designing a fictional race of bird people with human level intelligence. They have 4 limbs, a slightly curved upright stance and roughly 2m in height. Their hands and feet have 4 digits, 2 of those in the hands are opposable and dexterous enough for making/handling tools. Now, my main issue is designing the legs. As most are flightless, I planned for them to have adapted towards a more cursorial (adapted for running) path, and therefore wanted to make their legs similar to those of an ostrich. The issue is: I'd also like their legs to allow for some degree of climbing and, in extreme cases and much less relevant, swimming (their habitat is composed of large plains and decently large trees, located decently apart from one another). The issue? The leg design of ostriches is basically useless for matters other than standing up, running fast and for prolonged periods or kicking a lion/human to death.
In this comparison image, we can see how the ostrich's legs rely a lot on their tendons for the movement, which is what makes them so good at running but so bad for anything else (yellow is muscle, blue is tendons)
Given that my bird people already have arms to aid in the climbing (along with hand talons from their predatorial ancestors) , I'd think that having a zygodactyl foot, but with the 2 opposing digits located a bit higher up on the leg, as well as having slightly more flexible ankles capable of some rotation would be enough to solve the problem, but I'm having trouble implementing these on the leg, since I haven't found any good real life example of what I want. Can I make these legs capable of some climbing capability? I'm interested whether my assumption on the leg design is correct and possible to implement, or whether there are better ways to solve the climbing problem.
The plausibility of this part of their anatomy from an evolutionary perspective isn't necessary, but it'd be appreciated.