I stopped trusting Claude's code reviews, so I built a skill that puts my code on trial

Every time I asked Claude to review my branch, I got one of two answers: a cheerful "Looks good! 👍" or a vague list where I couldn't tell a real bug from a matter of taste. The model wants to please you. That's exactly the problem. So I built Tribunal — a Claude skill that reviews your diff adversarially, in stages, where the honest signal comes from agents fighting each other instead of one polite model. The idea: don't ask one model to be fair A single model told to "be critical"...