Turn LMs into SWE agents (and more)!
SWE-agent turns LMs (e.g. GPT-4) into agents that can 🐛 fix issues in real GitHub repositories, ⛳️ solve coding challenges, and 🔥 crack offensive cybersecurity challenges (EnIGMA mode).
-
Background & goals
Learn more about the project goals and academic research.
-
Installation
Three different ways to get started, including running installation-free in your browser.
-
Usage
Learn how to make the most out of SWE-agent.
-
Configuration
SWE-agent can be tweaked extensively without modifying the code.
-
Development
Dig into SWE-agent's code and build your own agent!
-
EnIGMA
EnIGMA turns SWE-agent into an offensive cybersecurity expert.
-
Changelog
See what's new in SWE-agent.