[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-72606":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":23,"hasPages":23,"topics":25,"createdAt":10,"pushedAt":10,"updatedAt":30,"readmeContent":31,"aiSummary":32,"trendingCount":16,"starSnapshotCount":16,"syncStatus":33,"lastSyncTime":34,"discoverSource":35},72606,"Agentless","OpenAutoCoder\u002FAgentless","OpenAutoCoder","Agentless🐱:  an agentless approach to automatically solve software development problems","",null,"Python",2068,234,20,51,0,4,7,23,12,29.11,"MIT License",false,"main",[26,27,28,29],"agent","artificial-intelligence","llm","software-development","2026-06-12 02:03:05","# 😺 Agentless\n\n\u003Cp align=\"center\">\n    \u003Ca href=\"https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.01489\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F📃-Arxiv-b31b1b?style=for-the-badge\">\u003C\u002Fa>\n    \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FOpenAutoCoder\u002FAgentless\u002Fblob\u002Fmaster\u002FLICENSE\">\u003Cimg src=\"https:\u002F\u002Fforthebadge.com\u002Fimages\u002Fbadges\u002Flicense-mit.svg\" style=\"height: 28px\">\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n    \u003Cbig>\u003Ca href=\"#-news\">😽News\u003C\u002Fa>\u003C\u002Fbig> |\n    \u003Cbig>\u003Ca href=\"#-setup\">🐈Setup\u003C\u002Fa>\u003C\u002Fbig> |\n    \u003Cbig>\u003Ca href=\"#-comparison\">🧶Comparison\u003C\u002Fa>\u003C\u002Fbig> | \n    \u003Cbig>\u003Ca href=\"#-artifacts\">🐈‍⬛Artifacts\u003C\u002Fa>\u003C\u002Fbig> |\n    \u003Cbig>\u003Ca href=\"#-citation\">📝Citation\u003C\u002Fa>\u003C\u002Fbig> |\n    \u003Cbig>\u003Ca href=\"#-acknowledgement\">😻Acknowledgement\u003C\u002Fa>\u003C\u002Fbig>\n\u003C\u002Fp>\n\n## 😽 News \n\n- *Dec 2nd, 2024*: We integrated Agentless with Claude 3.5 Sonnet to achieve 40.7% and 50.8% solve rate on SWE-bench lite and verified \n- *Oct 28th, 2024*: We just released OpenAutoCoder-Agentless 1.5! \n- *July 1st, 2024*: We just released OpenAutoCoder-Agentless 1.0! **Agentless** currently is the best open-source approach on SWE-bench lite with 82 fixes (27.3%) and costing on average $0.34 per issue.\n\n## 😺 About \n\n**Agentless** is an *agentless* approach to automatically solve software development problems. To solve each issue, **Agentless** follows a simple three phase process: localization, repair, and patch validation.\n- 🙀 **Localization**: Agentless employs a hierarchical process to first localize the fault to specific files, then to relevant classes or functions, and finally to fine-grained edit locations\n- 😼 **Repair**: Agentless takes the edit locations and samples multiple candidate patches per bug in a simple diff format\n- 😸 **Patch Validation**: Agentless selects the regression tests to run and generates additional reproduction test to reproduce the original error. Using the test results, Agentless re-ranks all remaining patches to selects one to submit\n\n## 🐈 Setup\n\nFirst create the environment \n\n```shell\ngit clone https:\u002F\u002Fgithub.com\u002FOpenAutoCoder\u002FAgentless.git\ncd Agentless\n\nconda create -n agentless python=3.11 \nconda activate agentless\npip install -r requirements.txt\nexport PYTHONPATH=$PYTHONPATH:$(pwd)\n```\n\n\u003Cdetails>\u003Csummary>⏬ Developer Setup\u003C\u002Fsummary>\n\u003Cdiv>\n\n```shell\n# for contribution, please install the pre-commit hook.\npre-commit install  # this allows a more standardized code style\n```\n\n\u003C\u002Fdiv>\n\u003C\u002Fdetails>\n\nThen export your OpenAI API key \n```shell\nexport OPENAI_API_KEY={key_here}\n```\n\nNow you are ready to run **Agentless** on the problems in SWE-bench! \n\n> [!NOTE]\n> \n> To reproduce the full SWE-bench lite experiments and follow our exact setup as described in the paper. Please see this [README](https:\u002F\u002Fgithub.com\u002FOpenAutoCoder\u002FAgentless\u002Fblob\u002Fmain\u002FREADME_swebench.md)\n\n## 🧶 Comparison\n\nBelow shows the comparison graph between **Agentless** and the best open-source agent-based approaches on SWE-bench lite\n\n\u003Cp align=\"center\">\n\u003Cimg src=\".\u002Fresources\u002Fcomparison_graph.png\" style=\"width:75%; margin-left: auto; margin-right: auto;\">\n\u003C\u002Fp>\n\n## 🐈‍⬛ Artifacts\n\nYou can download the complete artifacts of **Agentless** in our [v1.5.0 release](https:\u002F\u002Fgithub.com\u002FOpenAutoCoder\u002FAgentless\u002Freleases\u002Ftag\u002Fv1.5.0):\n- 🐈‍⬛ agentless_swebench_lite: complete Agentless run on SWE-bench Lite\n- 🐈‍⬛ agentless_swebench_verified: complete Agentless run on SWE-bench Verified\n- 🐈‍⬛ swebench_repo_structure: preprocessed structure information for each SWE-Bench problem\n\nYou can also checkout `classification\u002F` folder to obtain our manual classifications of SWE-bench-lite as well as our filtered SWE-bench-lite-*S* problems.\n\n## 📝 Citation\n\n```bibtex\n@article{agentless,\n  author    = {Xia, Chunqiu Steven and Deng, Yinlin and Dunn, Soren and Zhang, Lingming},\n  title     = {Agentless: Demystifying LLM-based Software Engineering Agents},\n  year      = {2024},\n  journal   = {arXiv preprint},\n}\n```\n\n> [!NOTE]\n> \n> The first two authors contributed equally to this work, with author order determined via [*Nigiri*](https:\u002F\u002Fsenseis.xmp.net\u002F?Nigiri)\n\n## 😻 Acknowledgement \n\n* [SWE-bench](https:\u002F\u002Fwww.swebench.com\u002F)\n* [Aider](https:\u002F\u002Fgithub.com\u002Fpaul-gauthier\u002Faider)\n* [SWE-bench-docker](https:\u002F\u002Fgithub.com\u002Faorwall\u002FSWE-bench-docker)\n","Agentless 是一个无需代理即可自动解决软件开发问题的工具。其核心功能包括通过三阶段流程（定位、修复和补丁验证）来处理每个问题，首先精确定位错误位置，然后生成多个候选补丁，并通过运行回归测试和额外的复现测试来选择最优补丁。该项目采用Python编写，利用了人工智能技术，特别是大型语言模型（LLM），以提高解决问题的效率与准确性。适合于需要自动化处理软件缺陷修复的场景，尤其是那些希望减少人工干预并加快开发周期的项目。根据项目文档，Agentless 在SWE-bench lite基准测试中表现优异，显示出其在成本效益方面的优势。",2,"2026-06-11 03:42:46","high_star"]