AI RESEARCH
Nautilus Compass: Black-box Persona Drift Detection for Production LLM Agents
arXiv CS.AI
•
ArXi:2605.09863v1 Announce Type: cross Production LLM coding agents drift over long sessions: they forget user-specified constraints, slip into mistakes the user already flagged, and confabulate prior agreements. White-box approaches such as persona vectors require model weights and so cannot be applied to closed APIs (Claude, GPT-4) that most users actually interact with. We present Nautilus Compass, a black-box persona drift detector and agent memory layer for production coding agents.