SWE-Agent
SWE-agent is a framework that lets a language model of your choice (e.g., GPT-4o or Claude Sonnet 4) autonomously use tools to fix issues in real GitHub repositories, find cybersecurity vulnerabilities, or perform custom tasks. It is designed for developers and researchers who want to automate software engineering tasks with AI. Note: the project is superseded by mini-swe-agent, which is recommended for new users.
🎬 SWE-Agent Team Interview - Agents, Programming, and Benchmarks! · Matthew Berman
✨ Key features
- State of the art on SWE-bench among open-source projects
- Free-flowing and generalizable: leaves maximal agency to the LM
- Configurable and fully documented via a single YAML file
- Made for research: simple and hackable by design
- Supports offensive cybersecurity mode (EnIGMA) for CTF challenges
🎯 Use cases
- Automatically fix bugs in real GitHub repositories
- Solve capture-the-flag cybersecurity challenges
- Benchmark language models on SWE-bench
- Perform custom coding tasks with language models
📦 Installation
🧰 Requirements: Requires Python and a language model API key (e.g., OpenAI or Anthropic). See documentation for details.
git clone https://github.com/SWE-agent/SWE-agent.git
cd SWE-agentconda create -n swe-agent python=3.10
conda activate swe-agentpip install -e .For more detailed installation instructions, see the documentation.
🚀 Usage
# Run SWE-agent on a GitHub issue
swe-agent run --model gpt-4o --issue https://github.com/repo/issue/123
For a full hello world example, see the documentation.
⚠️ Good to know
The project is superseded by mini-swe-agent, which is recommended for new users; SWE-agent is no longer the primary focus of development.
❓ FAQ
What is the relationship between SWE-agent and mini-swe-agent?
Mini-swe-agent has superseded SWE-agent and is recommended for new users. It matches SWE-agent's performance while being much simpler.
Can SWE-agent be used for cybersecurity tasks?
Yes, SWE-agent has a mode called EnIGMA for solving offensive cybersecurity (capture the flag) challenges, achieving state-of-the-art results on multiple benchmarks.
What language models are supported?
SWE-agent enables your language model of choice, such as GPT-4o or Claude Sonnet 4, to autonomously use tools.
Is SWE-agent open source?
Yes, it is released under the MIT license.
📊 Repository
🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.