
AI Summary
rockycode, a coding agent harness benchmarked against SWE-bench Verified, has been released on PyPI. The tool aims to evaluate and improve AI agents' ability to handle software engineering tasks.
From the source
A coding agent harness, benchmarked on SWE-bench Verified. amaze!
The full text couldn't be loaded here (the source may require a subscription).
View original at Pypi.org