Automated assessment is well-established within computer science courses but largely absent from human--computer interaction courses. Automating the assessment of human--computer interaction (HCI) is challenging because the coursework tends not to be computational but rather highly creative, such as designing and implementing interactive prototypes. We meet this challenge by developing an automarker for HCI 'MarCHIr' to assess key aspects of web prototypes: visual design, interactivity and accessibility. MarCHIr automatically compiles web prototypes and analyses them at the pixel level to assess designs using foundational Gestalt visual principles. While computer science student assessments are often personalised with permutations of numbers, we use random assignment of colours to personalise creative prototype assignments at the scale of hundreds of students. Finally, MarCHIr integrates industry accessibility checks to HCI student assessment. We share 2 years of case study data across two cohorts of university students and reflect on implications for the use of automated assessment within HCI tertiary education.