Agent Command Long-Term Duration: deterministic, auditable, fail-closed continuity for long-running AI agent missions.
Scoring was not performed