RIVA: Leveraging LLM Agents for Reliable Configuration Drift Detection

Researchers introduce RIVA, a multi-agent AI system that uses specialized verification agents and cross-validation to detect infrastructure configuration drift more reliably. The system improves accuracy from 27.3% to 50% when dealing with erroneous tool responses, addressing a critical reliability issue in cloud infrastructure management.

• →RIVA uses two specialized agents (verifier and tool generation) that collaborate through iterative cross-validation to detect infrastructure configuration drift.
• →The system addresses the problem of existing AI agents being vulnerable to erroneous tool responses that can cause missed drift detection or false alarms.
• →Testing shows RIVA recovers task accuracy from 27.3% to 50.0% when tools produce incorrect outputs.
• →Even without erroneous responses, RIVA improves baseline accuracy from 28% to 43.8%.
• →The research demonstrates that cross-validation of diverse tool calls enables more reliable autonomous infrastructure verification in production environments.