Pre-Mortem the Autonomous Change
Use a project pre-mortem to surface agent-run risks before implementation begins.
The risk move: make failure speak before the agent starts. A project pre-mortem works by assuming the future failure has already happened. In coding-agent work, that changes the conversation from vague caution to concrete prevention. Instead of asking whether the agent is good enough, ask what incident this exact task could create if the agent overreaches, misunderstands a hidden contract, or passes the wrong tests. The pre-mortem is especially useful because coding agents produce momentum. A large diff can appear in minutes, and the existence of a coherent patch can make the plan feel safer than it is. The pre-mortem…
Sign up free — one personalized lesson every day, matched to your role and goals.
Already have an account? Sign in