arxiv
PublishedJune 11, 2026 at 4:00 AM
Claw-SWE-Bench: A Benchmark for Evaluating OpenClaw-style Agent Harnesses on Coding Tasks
Publisher summary· verbatim
arXiv:2606.12344v1 Announce Type: cross Abstract: General-purpose agents such as OpenClaw are increasingly used as autonomous tool users, but their coding ability is difficult to measure under SWE-bench: a generic agent does not by itself satisfy the clean Docker workspace, patch, and prediction con
Stay posted· Newsletter
A 5-min weekly brief — top movers, price watch, story of the week.
Discussion
No replies yet. Be first.
Related coverage
More from ARXIV
arxivAnomalies in Multivariate Time Series Benchmarks Are Mostly Univariate1harxivLibra: Efficient Resource Management for Agentic RL Post-Training1harxivPianoKontext: Expressive Performance Rendering from Deadpan Context1harxivWhy Depth Matters in Parallelizable Sequence Models: A Lie Algebraic View1hThe Bubble Brief
WEEKLYRead AI insights every Tuesday — top movers, new releases, story of the week.
Originally published on arxiv ↗