SWE-Agent

Open-source agent that takes GitHub issues and attempts automated software fixes.

Visit site

Overview

SWE-Agent is an open-source software engineering agent from Princeton NLP. It takes a GitHub issue and tries to fix it using a language model selected by the user. The project can also be used for offensive cybersecurity and competitive programming challenges, according to its official repository. GitHub hosts the source code, documentation, issues, and contribution workflow.

Key features

  • GitHub issue input
  • Automated software-fix attempts
  • User-selected language model
  • Open-source repository
  • Support for coding challenges
  • Security research use noted by the project

Use cases

Show all 7Show fewer

Alternatives

  • CapyCloud coding-agent fleet for issue triage, code review, testing, and parallel engineering workai agent
  • ClineOpen-source coding agent available as an SDK, IDE extension, and command-line assistantai agent
  • ClippyOpen-source programming assistant that plans, writes, debugs, and tests codeai agent
  • Cody by SourcegraphAI coding assistant that uses code search and repository context to write, explain, and fix codeAI tool
  • ContinueOpen-source coding agent and developer tool for editor, terminal, and model workflowsAI tool
  • CursorAI coding agent and editor for writing, editing, and shipping software.ai agentFreemium

Last updated 2026-09-16