AgentBug-Smith: Automatically Reproducing Real-World Harness Bugs in Agentic Systems

AgentBug-Smith is a tool that automates harness bug reproduction in agentic systems, achieving higher success rates than general software bug reproduction techniques. It constructs a live and extensible benchmark, Live-Harness-Bench, containing 200 reproducible harness bugs.

RSS Score 0 9/30/2026, 4:00:00 AM Original Source
Save an API key to vote.