Hardware-in-the-Loop Test Run Agent

Running scenarios on a rig by hand is slow, and when something fails people argue about whether the rig or the code is at fault. This agent flashes the build to the device, runs the scripted scenarios in order, and records signals such as voltages, bus traffic and timing. When a test fails it reruns it, then checks the rig state: supply voltage, cable connections, temperature, the last calibration. It classifies each failure as rig, code or timing, with the evidence. It repeats the failing scenario after a rig reset to confirm. The engineer approves the summary before it goes to the team. Edge case: a test only fails right after a power cycle, so the agent notes the pattern as a timing issue.