arXiv cs.AI· Tao Yu, Xinyu Zhang, Qianqian Chen, Xiaoneng Xiang, Chia Kwangyang, Xingchen Huang, Ran Chen, Yangkai Ding, Zheng Wang, Yeo Boon Hong, Bingzheng Gan, Enrui Hu, Shuo Cheng, Deyang Li, Ruifeng Shi, Hongbo Wang, Qi Ye, Xuefeng Jin, Zhangchun Zhao·· 12 小时前AI 评分64
openJiuwen：面向长程编码智能体的开源可组合 harness
openJiuwen: Beyond Static Harnesses for Long-Horizon Coding Agents
AI 导读
openJiuwen 是一个面向长程编码智能体的开源 harness，提出结构可组合性与运行时自适应性两个设计目标。它通过共享执行底座和 Rail 能力组合，统一单智能体、委派子智能体与 Swarm Flow 的编排，并让语义诊断、执行结果等运行时证据动态影响上下文与任务控制。
正文 · 原文
Authors:Tao Yu, Xinyu Zhang, Qianqian Chen, Xiaoneng Xiang, Chia Kwangyang, Xingchen Huang, Ran Chen, Yangkai Ding, Zheng Wang, Yeo Boon Hong, Bingzheng Gan, Enrui Hu, Shuo Cheng, Deyang Li, Ruifeng Shi, Hongbo Wang, Qi Ye, Xuefeng Jin, Zhangchun Zhao
Abstract:Long-horizon coding agents operate over evolving repository states while increasingly relying on heterogeneous capabilities, delegated agents, and multi-agent coordination. These trends pose two complementary challenges for the agent harness. First, developers need to compose capabilities, reconfigure execution logic, and scale increasingly complex agent systems without repeatedly rebuilding orchestration. Second, complex coding tasks continuously produce new evidence---such as semantic diagnostics, execution outcomes, task progress, and changing context relevance---that should dynamically influence subsequent runtime decisions. We characterize these challenges as Structural Composability and Runtime Adaptivity. We present openJiuwen, an open-source harness designed for both developer composability and adaptive task execution. openJiuwen provides a shared execution substrate and Rail-based capability composition across single agents, delegated sub-agents, and Swarm Flow, enabling developers to construct sophisticated agent harnesses under common execution semantics. It further adapts framework-controlled runtime decisions around a fixed model policy, allowing evolving evidence to dynamically affect context, feedback, and task control toward successful completion. We systematically evaluate openJiuwen on SWE-bench Verified and Terminal-Bench 2.1, where it achieves 82.6% and 87.19%, respectively, exceeding the strongest selected official-leaderboard point estimates by 3.4 and 3.39 percentage points. These results show that openJiuwen achieves strong performance on complex coding tasks while providing a composable and adaptive harness design.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2608.27969 [cs.AI] |
| (or arXiv:2608.27969v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2608.27969 arXiv-issued DOI via DataCite |
Submission history
From: Bingzheng Gan [view email]
[v1]
Fri, 28 Aug 2026 06:31:38 UTC (6,007 KB)
[v2]
Thu, 5 Nov 2026 08:57:44 UTC (6,007 KB)
来源：arXiv cs.AI · arxiv.org