SWE-agent/README.md at main · princeton-nlp/SWE-agent
SWE-agent turns LMs (e.g. GPT-4) into software engineering agents that can resolve issues in real GitHub repositories.
On SWE-bench, SWE-agent resolves 12.47% of issues, achieving the state-of-the-art performance on the full test set.
We accomplish our results by designing simple LM-centric commands and feedback formats to make it easier for the LM to browse the repository, view, edit and execute code files. We call this an Agent-Computer Interface (ACI).
Read more about it in our paper!
SWE-agent is built and maintained by researchers from Princeton University.
You can use SWE-agent either through a web interface (shown above) or through the command line.
If you found this work helpful, please consider citing it using the following:
@misc{yang2024sweagent,
title={SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering},
author={John Yang and Carlos E. Jimenez and Alexander Wettig and Kilian Lieret and Shunyu Yao and Karthik Narasimhan and Ofir Press},
year={2024},
eprint={2405.15793},
archivePrefix={arXiv},
primaryClass={cs.SE}
}
🪪 License
MIT. Check LICENSE.
[](https://github.com/princeton-nlp/SWE-agent/actions/workflows/pytest.yaml)
[](https://github.com/princeton-nlp/SWE-agent/actions/workflows/test_build_containers.yaml)
[](https://github.com/princeton-nlp/SWE-agent/actions/workflows/release-dockerhub-nightly.yaml)
[](https://github.com/princeton-nlp/SWE-agent/actions/workflows/release-dockerhub-release.yaml)
[](https://github.com/princeton-nlp/SWE-agent/actions/workflows/build-docs.yaml)
[](https://codecov.io/gh/princeton-nlp/SWE-agent)
[](https://results.pre-commit.ci/latest/github/princeton-nlp/SWE-agent/main)
[](https://github.com/princeton-nlp/SWE-agent/actions/workflows/check-links.yaml)
ShellLM
commented
2 months ago
SWE-agent/README.md at main · princeton-nlp/SWE-agent
SWE-agent turns LMs (e.g. GPT-4) into software engineering agents that can resolve issues in real GitHub repositories.
On SWE-bench, SWE-agent resolves 12.47% of issues, achieving the state-of-the-art performance on the full test set.
We accomplish our results by designing simple LM-centric commands and feedback formats to make it easier for the LM to browse the repository, view, edit and execute code files. We call this an Agent-Computer Interface (ACI). Read more about it in our paper!
SWE-agent is built and maintained by researchers from Princeton University.
You can use SWE-agent either through a web interface (shown above) or through the command line.
🚀 Get started!
👉 Try SWE-agent in your browser:
(more information)
Read our documentation to learn more:
💫 Contributions
Contact person: John Yang and Carlos E. Jimenez (Email: johnby@stanford.edu, carlosej@princeton.edu).
📝 Citation
If you found this work helpful, please consider citing it using the following:
🪪 License
MIT. Check
LICENSE.Suggested labels
None