Accuracy of an auto assessment tool. Compare students' mark from the tool and from the teachers.

In my thesis I used an auto assessment tool to automatically put marks on students during an educational activity. The activity comprises of 10 stages and 1 mark is given to students after the completion of each stage. In addition, I used 97 teachers to put marks, as well, to the same students depending on how many stages they successful completed - as the tool did after the completion of the activity. I want to prove the accuracy of the tool by comparing the marks of the tool for each student with the marks for each one of the teachers they put marks. Hope I described my problem well.
I want to mention that I found the inter rater agreement as a way but I am not sure if it compares one the one hand the specific marks of the tool and on the other the specific marks from each teacher.
Thanking you in advance