I think the most natural answers are the contributions to science by Popper and Kuhn. They developed what I would loosely consider "ground rules" which ensure science stays on point. Popper advanced the concept of falsification, which provides a key piece of humility to science: science never proves anything. It merely disproves theories which are inconsistent with reality. Using the terminology of the scientific method as taught in schools, we never prove the alternate hypothesis to be true. We merely reject the null hypothesis because it is statistically unlikely to be true.
Kuhn built on this and added a social aspect to science. Kuhn's theories claim that science runs through two phases. In one phase, there is a strong effort to demonstrate that the current theories work, and refine them into better theories. In the other phase, completely new theories are created to replace the old theories entirely. I would argue it's Kuhn's work that permits one to talk about "scientific truths," because he is the one who defined the social constructs which capture that.
Beyond that, I think most of the time it's the philosophy providing a stable foundation for the science. I'd say its analogous to the famous ad by BASF, a chemical company:
At BASF, we don't make a lot of the products you buy, we make a lot of the products you buy better.
And sometimes, it's really hard to tell the philosophy apart from the science. If you read Dr. Wilson's testimony before Congress for the construction of Fermilab, I find it's not easy to see where the boundary between the science and the philosophy lies. In such cases, what you describe is a demarcation problem. You have two things intimately entwined together, and you ask what the contribution of one half is. But it wasn't clear where to draw the line in the first place. None the less, I think there are reasonable arguments to be made for the contributions of philosophy to furthering science. Philosophy shows science its place, as all of science's arguments fit square in the philosophical concept of empiricism. Given that there's a known demarcation problem separating science from pseudo-science, I think providing some bounds on science is a value added effort. At the very least, philosophy can give you a framework within which you can prove that some concepts, like God or freewill, are simply out of scope for science. That makes science more efficient, as it isn't busy wasting its time on things like that.