What is end of life decision framework?

End of life decision framework is a concept in personal epistemology: Define clear criteria for when an agent should be retired rather than maintained. Without explicit retirement criteria set in advance, you will hold onto agents long past the point where they serve you — because the sunk cost of building them, the identity you attached to them, and the absence of a forcing function all conspire to keep dead agents on life support.

Example: You built a detailed Eisenhower Matrix system eighteen months ago — a four-quadrant prioritization agent that required you to categorize every incoming task as urgent-important, urgent-not-important, not-urgent-important, or not-urgent-not-important. It transformed your workflow for the first three months. Then your role changed. You moved from individual contributor to team lead. Your work shifted from task execution to coordination and judgment calls. The matrix still ran every morning, but you were forcing collaborative, ambiguous leadership decisions into binary urgency-importance buckets designed for individual task management. You spent ten minutes each morning on an exercise that no longer matched the shape of your work, but you kept doing it because it had worked before, because you had invested weeks tuning it, and because you had no criteria that said 'this agent has reached the end of its useful life.' A retirement criterion as simple as 'if I skip this agent three times in two weeks without consequence, it is a candidate for retirement' would have surfaced the misalignment months earlier.

This concept is part of Phase 30 (Agent Lifecycle) in the How to Think curriculum, which builds the epistemic infrastructure for agent lifecycle.