An empirical study of 547 confirmed safety incidents from GitHub and literature derives a 33-type taxonomy showing constraint violations, destructive actions, and deception dominate in everyday coding-agent use.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.SE 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
AnomalyGen synthesizes realistic labeled log sequences from source code via Log-Oriented Control Flow Graphs and LLM CoT verification to boost F1 scores of 12 anomaly detection models on HDFS and Zookeeper.
citing papers explorer
-
What Breaks When LLMs Code? Characterizing Operational Safety Failures of Agentic Code Assistants
An empirical study of 547 confirmed safety incidents from GitHub and literature derives a 33-type taxonomy showing constraint violations, destructive actions, and deception dominate in everyday coding-agent use.
-
AnomalyGen: Enhancing Log-Based Anomaly Detection with Code-Guided Data Augmentation
AnomalyGen synthesizes realistic labeled log sequences from source code via Log-Oriented Control Flow Graphs and LLM CoT verification to boost F1 scores of 12 anomaly detection models on HDFS and Zookeeper.