1 pointby DAVI_RMX4 hours ago1 comment
  • DAVI_RMX4 hours ago
    I built NEXO, a governance-first execution runtime for AI agents.

    The idea is simple: instead of evaluating agents only by task performance, NEXO treats agent execution as a controlled system with policy gates, risk thresholds, and runtime constraints.

    The runtime classifies behavior into regimes (healthy, defensive, unstable) and measures operational stability under different environment conditions.

    In experiments, a sharp stability frontier appears: - below a minimum risk threshold → agents collapse - above the threshold → stability emerges - increasing uncertainty collapses stability again - no gradual degradation band observed (phase-transition behavior)

    The repository includes: - governed execution runtime - reproducible experiment harness - stability frontier analysis - policy and capability controls

    This is early research. Feedback and criticism welcome.