Automatically Fix Github Issues In Real Repositories

SWE-agent lets a chosen LM autonomously use tools to fix GitHub issues in real repos, hitting state-of-the-art open-source SWE-bench results. It also supports offensive-security (EnIGMA) and competitive coding tasks, configured via YAML. Note: development has moved to mini-SWE-agent.

AI Coding | Python · swe-bench · software-engineering