FLARE: Agentic Coverage-Guided Fuzzing for LLM-Based Multi-Agent Systems
This addresses reliability issues in automated complex workflows using LLM-based multi-agent systems, representing a novel method for a known bottleneck.
The paper tackles the problem of frequent failures in Multi-Agent LLM Systems due to non-deterministic behavior and complex interactions, presenting FLARE, a testing framework that achieves 96.9% inter-agent and 91.1% intra-agent coverage and uncovers 56 previously unknown failures.
Multi-Agent LLM Systems (MAS) have been adopted to automate complex human workflows by breaking down tasks into subtasks. However, due to the non-deterministic behavior of LLM agents and the intricate interactions between agents, MAS applications frequently encounter failures, including infinite loops and failed tool invocations. Traditional software testing techniques are ineffective in detecting such failures due to the lack of LLM agent specification, the large behavioral space of MAS, and semantic-based correctness judgment. This paper presents FLARE, a novel testing framework tailored for MAS. FLARE takes the source code of MAS as input and extracts specifications and behavioral spaces from agent definitions. Based on these specifications, FLARE builds test oracles and conducts coverage-guided fuzzing to expose failures. It then analyzes execution logs to judge whether each test has passed and generates failure reports. Our evaluation on 16 diverse open-source applications demonstrates that FLARE achieves 96.9% inter-agent coverage and 91.1% intra-agent coverage, outperforming baselines by 9.5% and 1.0%. FLARE also uncovers 56 previously unknown failures unique to MAS.