The Agent Security League scores coding agents on 200 tasks drawn from 108 open-source Python projects covering 77 vulnerability classes, and extends Carnegie Mellon's SusVibes benchmark. A result counts as functional if it passes the…
In the News: October 23, 2026 (Extra 3)
The best coding agent on Endor Labs' security benchmark passes the security tests on 37.4% of tasks, against 87.2% on functional tests.
Extra edition
Machine-readable
Download Markdown