Swe Agent

SWE-agent is an open-source autonomous software engineering agent developed by researchers at Princeton University and Stanford University. It enables large language models (e.g., GPT-4, Claude) to autonomously interact with GitHub repositories—navigating codebases, identifying bugs, and submitting fixes—via a custom Agent-Computer Interface (ACI) optimized for LLM use. It also supports offensive cybersecurity tasks and competitive coding challenges, and achieved state-of-the-art results on the SWE-bench benchmark. The project was published at NeurIPS 2024 and has since been succeeded by mini-swe-agent, a simpler and equally performant successor.