Harness-Bench: Evaluating Agent Harnesses across Models and Realistic Workflows
arXiv:2605.27922v2 Announce Type: replace Abstract: LLM agents are increasingly deployed as executable systems that use tools, modify workspaces, and produce concrete artifacts. In such workflows, performance depends not only on the base model, but also on the harness: the system layer that manages context, tools, state, constraints, permissions, tracing, and recovery. However, existing benchmarks typically abstract away execution, compare complete agent systems, or hold the harness fixed, maki…