An Approach to Checking Correctness for Agentic Systems
This provides a systematic monitoring approach for AI agent reliability in critical applications, though it represents an incremental adaptation of temporal logic techniques from hardware verification.
The paper tackles the problem of detecting errors in LLM-based agentic systems by developing a temporal expression language that monitors agent behavior sequences rather than text outputs, successfully identifying behavioral regressions when smaller models caused improper tool sequencing and coordination failures.
This paper presents a temporal expression language for monitoring AI agent behavior, enabling systematic error-detection of LLM-based agentic systems that exhibit variable outputs due to stochastic generation processes. Drawing from temporal logic techniques used in hardware verification, this approach monitors execution traces of agent tool calls and state transitions to detect deviations from expected behavioral patterns. Current error-detection approaches rely primarily on text matching of inputs and outputs, which proves fragile due to the natural language variability inherent in LLM responses. The proposed method instead focuses on the sequence of agent actions -- such as tool invocations and inter-agent communications -- allowing verification of system behavior independent of specific textual outputs. The temporal expression language provides assertions that capture correct behavioral patterns across multiple execution scenarios. These assertions serve dual purposes: validating prompt engineering and guardrail effectiveness during development, and providing regression testing when agents are updated with new LLMs or modified logic. The approach is demonstrated using a three-agent system, where agents coordinate to solve multi-step reasoning tasks. When powered by large, capable models, all temporal assertions were satisfied across many test runs. However, when smaller models were substituted in two of the three agents, executions violated behavioral assertions, primarily due to improper tool sequencing and failed coordination handoffs. The temporal expressions successfully flagged these anomalies, demonstrating the method's effectiveness for detecting behavioral regressions in production agentic systems. This approach provides a foundation for systematic monitoring of AI agent reliability as these systems become increasingly deployed in critical applications.