Computer laboratories are an important facility in supporting the technology-based learning process in schools, but they often face the problem of hardware and software malfunctions that can hinder academic activities. The limited number of technicians in the school environment causes the damage diagnosis process to often be slow and less systematic. This research aims to design and build an embedded system-based computer damage diagnosis expert system that can be used as a troubleshooting tool in school laboratories. The method used is a system engineering approach (Research and Development) with the stages of needs analysis, design, implementation, and testing. The expert system's inference mechanism uses the rule-based Forward Chaining method implemented on the ESP32 microcontroller platform. The test results of 30 crashes showed a system accuracy rate of 86.67% with an average response time of less than 2 seconds. The system is able to provide early diagnosis and solution recommendations quickly and consistently. Thus, the system developed is effective as a decision support tool in the process of identifying computer damage and contributes to improving the efficiency of school laboratory management.