Researchers have introduced GH-AWTrace, a comprehensive dataset designed to capture the execution logs of agentic workflows on GitHub. This dataset links over 25,000 verified workflows with millions of run records and execution log archives, providing detailed insights into the processes behind automated repository tasks. The logs reveal critical information such as tool denials, agent failures, and inter-agent delegation, enabling further research into execution anomalies and the effectiveness of AI agents. AI
IMPACT Enables deeper research into the execution and failure modes of AI agents in software development workflows.
RANK_REASON The cluster contains a research paper detailing a new dataset for studying AI agentic workflows. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Hugging Face Daily Papers →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →