Pre-mortem prompt: stress-test a feature before you ship it

It's [6 months] after we shipped [feature name]. It failed. Not a small stumble, it's being rolled back, quietly killed, or has become a liability we regret. Assume this as fact and work backwards. Here's the feature: [1-3 sentence description, who it's for, and the core value prop]. Walk me through what went wrong across these angles: Adoption: why didn't people use it, or use it the way we hoped? What did we assume about user motivation or behavior that turned out to be wrong? Discoverability & onboarding: how did users fail to find it, understand it, or get to first value? Technical & operational: what broke, degraded, or didn't scale? What edge cases, data issues, or integration failures bit us? Second-order effects: what did it break elsewhere? Did it cannibalize another feature, confuse the product's story, create support load, or invite abuse? Business & metrics: did it move the wrong number, or move the right number the wrong way? Did we measure the wrong thing and declare false success? The thing nobody said out loud: what concern did people privately have that we talked ourselves out of? For each failure mode: state the specific mechanism (not "users didn't like it" but why), rate its likelihood (high/med/low) and how early we'd see the warning signs, and give one concrete thing we could do now to prevent or de-risk it. Be blunt. I want the uncomfortable version, not the diplomatic one.