Back to Rankings返回排行榜
Top 100 · Software Engineering Agents前 100 · 软件工程智能体
30 repositories sorted by software engineering agents 按 软件工程智能体 排序，共 30 个仓库
| # | Repository仓库 | Stars | Forks | Language语言 | Issues | Description描述 | Last Commit最后提交 |
|---|---|---|---|---|---|---|---|
| 1 | SWE-agent SWE-agent | 20.5k | 2.2k | Python | 55 | SWE-agent takes a GitHub issue and tries to automatically fix it, using your LM of choice. It can also be employed for offensive cybersecurity or competitive coding challenges. [NeurIPS 2024] SWE-agent 接收 GitHub 问题并尝试使用您选择的 LM 自动修复它。它还可以用于进攻性网络安全或竞争性编码挑战。 [神经IPS 2024] | 2026-10-06 |
| 2 | mini-swe-agent SWE-agent | 8.4k | 1.1k | Python | 36 | The 100 line AI agent that solves GitHub issues or helps you in your command line. Radically simple, no huge configs, no giant monorepo—but scores >74% on SWE-bench verified!100 行 AI 代理可以解决 GitHub 问题或在命令行中为您提供帮助。极其简单，没有巨大的配置，没有巨大的单一存储库，但在 SWE-bench 验证上得分 >74%！ | 2026-10-06 |
| 3 | CUAWright microsoft | 6.0k | 386 | Python | 15 | A simple SWE style browser+desktop agent framework that achieves SOTA results on long horizon web tasks. | 2026-10-06 |
| 4 | rllm rllm-org | 5.9k | 630 | Python | 79 | Democratizing Reinforcement Learning for LLMsError 500 (Server Error)!!1500.That’s an error.There was an error. Please try again later.That’s all we know. | 2026-10-09 |
| 5 | agentic-harness-engineering china-qijizhifeng | 915 | 103 | Python | 5 | Official AHE code — Agentic Harness Engineering: observability-driven automatic evolution of coding-agent harnesses (concurrent w/ meta-harness). NexAU-AHE reaches 84.7% ± 2.1 pass@1 on Terminal-Bench 2 (GPT-5.5). Lifts GPT-5.4 69.7→77.0% over 10 iters, beats Codex/ACE/Training-Free GRPO; frozen harness transfers to SWE-bench-Verified.官方 AHE 代码 — 代理线束工程：可观察性驱动的编码代理线束自动演化（与元线束并发）。 NexAU-AHE 在 Terminal-Bench 2 (GPT-5.5) 上达到 84.7% ± 2.1 pass@1。在 10 迭代中提升 GPT-5.4 69.7→77.0%，击败 Codex/ACE/Training-Free GRPO；冷冻线束转移至 SWE 工作台验证。 | 2026-08-03 |
| 6 | augment-swebench-agent augmentcode | 885 | 157 | Python | 7 | The #1 open-source SWE-bench Verified implementation#1 开源 SWE-bench 已验证实施 | 2026-08-21 |
| 7 | SWE-smith SWE-bench | 798 | 136 | Python | 17 | [NeurIPS 2025 D&B Spotlight] Scaling Data for SWE-agents[NeurIPS 2025 D&B 聚焦] SWE 代理的扩展数据 | 2026-10-05 |
| 8 | SWE-Gym SWE-Gym | 749 | 46 | Jupyter Notebook | 9 | Code for Paper: Training Software Engineering Agents and Verifiers with SWE-Gym [ICML 2025]论文代码：使用 SWE-Gym 培训软件工程代理和验证者 [ICML 2025] | 2025-07-29 |
| 9 | swe-agent langtalks | 642 | 130 | Python | 1 | 🤖 AI-powered software engineering multi-agent system with researcher and developer agents that automate code implementation through intelligent planning and execution. Built with LangGraph multi-agent workflows🤖 人工智能驱动的软件工程多代理系统，具有研究人员和开发人员代理，可通过智能规划和执行自动执行代码。使用 LangGraph 多代理工作流程构建 | 2026-03-28 |
| 10 | SWE-ReX SWE-agent | 618 | 129 | Python | 35 | Sandboxed code execution for AI agents, locally or on the cloud. Massively parallel, easy to extend. Powering SWE-agent and more. 用于本地或云端人工智能代理的沙盒代码执行。大规模并行，易于扩展。为 SWE 代理等提供支持。 | 2026-10-06 |
| 11 | mycoder bhouston | 567 | 52 | TypeScript | 56 | Simple to install, powerful command-line based AI agent system for coding.安装简单、功能强大、基于命令行的 AI 代理系统用于编码。 | 2026-01-07 |
| 12 | RepoMaster QuantaAlpha | 552 | 72 | Python | 3 | RepoMaster: The open-source AI agent that masters GitHub. It turns any code repository into a powerful tool, achieving a new level of autonomous task-solving. An open alternative to Claude-Code.RepoMaster：掌握 GitHub 的开源 AI 代理。它将任何代码存储库变成一个强大的工具，实现自主任务解决的新水平。克劳德代码的开放替代方案。 | 2025-11-05 |
| 13 | SWE-bench_Pro-os scaleapi | 541 | 101 | Python | 34 | SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?SWE-Bench Pro：人工智能代理能否解决长期软件工程任务？ | 2026-09-22 |
| 14 | AI-Coding-Style-Guides lidangzzz | 496 | 50 | JavaScript | 2 | A set of coding style guidelines for Vibe Coding or SWE-Agents that maximize efficiency and improve human readability.一套适用于 Vibe Coding 或 SWE-Agents 的编码风格指南，可最大限度地提高效率并提高人类可读性。 | 2025-09-15 |
| 15 | live-swe-agent OpenAutoCoder | 466 | 47 | N/A | 7 | Live-SWE-agent: live, runtime self-evolving software engineering agentLive-SWE-agent：实时、运行时自我进化的软件工程代理 | 2026-01-19 |
| 16 | agent-sweep-engine phoenix-zhou | 391 | 3 | Python | 0 | Built on **Agentic RAG** (Agent-driven Retrieval-Augmented Generation) technology, it not only accurately answers pre-sales and after-sales questions but also generates personalized usage reports and optimization suggestions by deeply analyzing device data.Error 500 (Server Error)!!1500.That’s an error.There was an error. Please try again later.That’s all we know. | 2026-07-29 |
| 17 | joycode-agent jd-opensource | 345 | 20 | Python | 1 | Repository-level Repair Agent Based on SWE-Bench—JoyCode Agent基于SWE-Bench的存储库级修复代理—JoyCode Agent | 2025-10-11 |
| 18 | R2E-Gym R2E-Gym | 339 | 68 | Python | 10 | [COLM 2025] Official repository for R2E-Gym: Procedural Environment Generation and Hybrid Verifiers for Scaling Open-Weights SWE Agents[COLM 2025] R2E-Gym 官方存储库：用于扩展开放权重 SWE 代理的程序环境生成和混合验证器 | 2025-07-13 |
| 19 | PRarena aavetis | 300 | 24 | HTML | 7 | This repo tracks the opened and merged PRs by the top SWE coding agents by OpenAI, GitHub, and others. Updates regularly.该存储库跟踪 OpenAI、GitHub 等顶级 SWE 编码代理打开和合并的 PR。定期更新。 | 2026-10-10 |
| 20 | SE-Agent JARVIS-Xs | 288 | 29 | Python | 5 | SE-Agent is a self-evolution framework for LLM Code agents. It enables trajectory-level evolution to exchange information across reasoning paths via Revision, Recombination, and Refinement, expanding the search space and escaping local optima. On SWE-bench Verified, it achieves SOTA performanceSE-Agent是LLM代码代理的自我进化框架。它使轨迹级进化能够通过修正、重组和细化跨推理路径交换信息，从而扩展搜索空间并逃离局部最优。在 SWE-bench Verified 上，它实现了 SOTA 性能 | 2025-09-23 |
| 21 | GitTaskBench QuantaAlpha | 259 | 20 | Python | 1 | Repo-level benchmark for real-world Code Agents: from repo understanding → env setup → incremental dev/bug-fixing → task delivery, with cost-aware α metric.现实世界代码代理的回购级基准：从回购理解→环境设置→增量开发/错误修复→任务交付，具有成本感知的α指标。 | 2025-09-22 |
| 22 | remote-swe-agents aws-samples | 244 | 50 | TypeScript | 19 | Autonomous SWE agent working in the cloud!在云端工作的自主 SWE 代理！ | 2026-10-06 |
| 23 | SWE-CI SKYLENAGE-AI | 181 | 20 | Python | 1 | SWE-CI: Evaluating Agent Capabilities in Maintaining Codebases via Continuous IntegrationSWE-CI：通过持续集成评估代理维护代码库的能力 | 2026-06-10 |
| 24 | DM-Code-Agent hwfengcs | 153 | 15 | Python | 1 | Local-first, auditable Python code agent. Ships its own 30-task hidden-test benchmark plus SWE-bench Verified scored by the official harness -- every number reproducible, negative results included.本地优先、可审计的 Python 代码代理。提供自己的 30 项隐藏测试基准测试以及由官方工具评分的 SWE-bench Verified - 每个数字都可重现，包括负面结果。 | 2026-08-27 |
| 25 | AgentKernelArena AMD-AGI | 121 | 14 | Python | 8 | AgentKernelArena provides an end-to-end siloed-benchmarking environment where different LLM-powered agents—such as Cursor Agent, Claude Code, Codex, SWE-agent, and GEAK—can be evaluated side-by-side on the same GPU kernel tasks, using objective and reproducible metrics.AgentKernelArena 提供了一个端到端的孤立基准测试环境，其中不同的 LLM 支持的代理（例如 Cursor Agent、Claude Code、Codex、SWE-agent 和 GEAK）可以使用客观且可重复的指标在相同的 GPU 内核任务上并行评估。 | 2026-10-10 |
| 26 | ToM-SWE OpenHands | 120 | 17 | Python | 0 | The theory of mind module for the SWE agentSWE 代理的心理理论模块 | 2026-08-03 |
| 27 | claw-swe-bench TokenRhythm | 110 | 10 | Python | 1 | Unified adapter framework for evaluating agent harnesses (claws) on SWE-bench用于在 SWE-bench 上评估代理线束（爪子）的统一适配器框架 | 2026-09-28 |
| 28 | SWE-Master RUCAIBox | 108 | 7 | HTML | 4 | SWE-Master: Unleashing the Potential of Software Engineering Agents via Post-TrainingSWE-Master：通过后期培训释放软件工程代理的潜力 | 2026-02-28 |
| 29 | agent-base giraffe-tree | 104 | 7 | HTML | 0 | Agent Base is a source-level research project on coding agents. It compares Codex CLI, OpenCode, Gemini CLI, Kimi CLI, and SWE-agent across agent loops, tools, MCP integration, context/memory handling, UI flows, web architecture, and safety controls.Agent Base 是一个关于编码代理的源码级研究项目。它比较了 Codex CLI、OpenCode、Gemini CLI、Kimi CLI 和 SWE-agent 的代理循环、工具、MCP 集成、上下文/内存处理、UI 流程、Web 架构和安全控制。 | 2026-08-20 |
| 30 | Active-SWE XLearning-SCU | 102 | 1 | Python | 0 | Pytorch Implementation of Active-SWE: Benchmarking Coding Agents for Proactive Bug Fixing without Issue ReportsActive-SWE 的 Pytorch 实现：对编码代理进行基准测试，无需问题报告即可主动修复错误 | 2026-08-14 |
No repositories match your search
没有匹配的仓库