Most scientists are 'tools' for years at the beginning of their careers. The project is steered by someone who likely has years more experience in a field, and 99% of the time knows better than you possibly could how to get results for a certain question. At this stage in our young lives, there is no shame in doing bench work with only a minor cerebral input into the project. With time and proficiency in whatever techniques, you will move on and gain and more responsibility, but that takes years.
You're right, the PI or the Post-Doc is going to have much larger input than any undergrad ever will but the point about "not being a tool" is still very valid. When I interviewed people to replace me at the research job I had before starting my MD/PhD, there was one candidate who really exemplified the idea that they were just a tool (I prefer to call them "a set of hands"). The guy had listed that he had experience in a genetics lab, so I asked him to tell me about it:
Him: "We did PCR."
Me: "Ok, but what were you looking at?"
Him: "Well we were running PCR on different genes."
Me: "...What genes were you looking at?"
Him: "Well I was assisting a graduate student so we were looking at the genes he was studying."
I'll spare you the rest. Needless to say this kid had no idea what was going on. I'm sure if you handed him the materials he could efficiently perform a PCR but the fact is he was just an extra set of hands for some grad student. He had zero input and in fact did not even know what he was actually doing. He could not explain what or why he did anything. Obviously he didn't get the job.
As an undergrad, your goal is to learn how research is performed. It's ok if every decision is made/approved by someone else but you have to be involved in the process. You should at least be able to tell me why every decision is made and even if you would be wrong, you should be able to at least give a somewhat reasonable answer to questions about designing/furthering the project. Real scientists do take years to train but by the end of undergrad you should be able to handle small scale project planning (e.g. figuring out the next step based on your results). If you are at this level, you will certainly not sound like the guy above.
A good LOR from your prof will do way more than a paper. Papers are certainly nice but there is so much out of your control that goes into it that it's not what defines an experience. For example I worked for 2 months as a data analyst using data from a clinical trial. The paper to publish that clinical trial took 6 years from the start of the trial until the paper was ready to submit, but the epidemiology work I did on data from the entry questionnaires generated 2 papers in 2 months. Since the 2 papers I was on used data from the clinical trial, they couldn't be submitted until the clinical trial was accepted. There was some administrative issue with the clinical trial such that the 2 papers with my name on it didn't get submitted until recently (5 years after we finished drafting them).
As long as you sound like you were involved in the planning and analysis of the research, the experience will help with your app. What field or type you do doesn't really matter.