SWE-agent

SWE-agent takes a GitHub issue and tries to automatically fix it, using your LM of choice. It can also be employed for offensive cybersecurity or competitive coding challenges. [NeurIPS 2024]

SWE-agent is an open-source coding & software engineering AI agent maintained by SWE-agent. It ranks #75 on The Agentic Leaderboard with an overall score of 83.3 / 100, scoring 97.4% on reliability and 71.2% on tool selection quality. The project has 19,589 GitHub stars.

Add this badge to your README

[![Agentic Leaderboard](https://www.theagenticleaderboard.com/badges/swe-agent.svg)](https://www.theagenticleaderboard.com/agent/swe-agent)