Towards Context-Aware Cloud Anomaly Detection: Benchmark, Agent, and Analysis

Anomaly detection in cloud environments remains critical and challenging. Existing context-level benchmarks typically focus on either metrics or logs and often lack reliable annotation, while most detection methods emphasize point anomalies within a single modality, overlooking contextual signals and limiting real-world applicability. Constructing a benchmark for context anomalies that combines metrics and logs is difficult: reproducing anomalous scenarios on real servers is often infeasible or harmful, while generating synthetic data introduces the challenge of maintaining cross-modal consistency. We introduce CLOUDANOBENCH, a large-scale benchmark for context anomalies in cloud environments, comprising 28 anomalous scenarios and 16 deceptive normal scenarios, with 1,252 labeled cases and roughly 200,000 log and metric entries. Compared with prior benchmarks, CLOUDANOBENCH exhibits higher ambiguity and greater difficulty, on which prior machine learning methods and vanilla LLM prompting perform poorly. To demonstrate its utility, we further propose CLOUDANOAGENT, an LLM-based agent enhanced by symbolic verification. CLOUDANOAGENT adopts a Signal Detection and Contextual Verification mechanism that combines metric-based anomaly screening with log-based contextual reasoning. This agent achieves substantial improvements in both anomaly detection and scenario identification on CLOUDANOBENCH, and shows strong generalization to existing datasets. Together, CLOUDANOBENCH and CLOUDANOAGENT provide a benchmark and agent framework for context-aware cloud anomaly detection.