openJiuwen: Beyond Static Harnesses for Long-Horizon Coding Agents
On the map: LLM agents and tool use → Software engineering coding agents
Abstract (arXiv)
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.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use. Builds on 21 of its 23 references that are in the map. First posted on arXiv 28 Aug 2026.
Builds on
21 of 23 references.
Closest ancestors · 12
Same direction, newest first
- 2026SWE-Pruner: Self-Adaptive Context Pruning for Coding Agents23 Jan
- 2025Live-SWE-agent: Can Software Engineering Agents Self-Evolve on the Fly?17 Nov
- 2025Understanding Code Agent Behaviour: An Empirical Study of Success and Failure Trajectories31 Oct
- 2025Lingxi: Repository-Level Issue Resolution Framework Enhanced by Procedural Knowledge Guided Scaling13 Oct
- 2025Reducing Cost of LLM Agents with Trajectory Reduction28 Sep
- 2025Trae Agent: An LLM-based Agent for Software Engineering with Test-time Scaling31 Jul
- 2025Understanding Software Engineering Agents: A Study of Thought-Action-Result Trajectories23 Jun
- 2024OpenHands: An Open Platform for AI Software Developers as Generalist Agents23 Jul
- 2024Agentless: Demystifying LLM-based Software Engineering Agents1 Jul
- 2024SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering6 May
- 2024RepairAgent: An Autonomous, LLM-Based Agent for Program Repair25 Mar
- 2024Executable Code Actions Elicit Better LLM Agents1 Feb
Other references · 9
- 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,193
- 2026Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfacescited by 515
- 2024AutoCodeRover: Autonomous Program Improvementcited by 272
- 2024Alibaba LingmaAgent: Improving Automated Issue Resolution via Comprehensive Repository Explorationcited by 58
- 2025The Complexity Trap: Simple Observation Masking Is as Efficient as LLM Summarization for Agent Context Managementcited by 40
- 2026SWE-AGILE: A Software Agent Framework for Efficiently Managing Dynamic Reasoning Contextcited by 5
- 2025P3: Prompts Promote Promptingcited by 3
- 2026Swarm Skills: A Portable, Self-Evolving Multi-Agent System Specification for Coordination Engineeringcited by 3
- 2026Turning Interaction History into Execution State: A Runtime Layer for Long-Horizon Coding Agentscited by 2
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026OpenCollab: A Multi-Agent Coding Framework with Programmable Collaboration and Controllable Runtime
- 2026Code as Agent Harness
- 2026SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery
- 2026Inside the Scaffold: A Source-Code Taxonomy of Coding Agent Architectures
- 2026Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses
- 2026EvoCode-Bench: Evaluating Coding Agents in Multi-Turn Iterative Interactions
- 2025NL2Repo-Bench: Towards Long-Horizon Repository Generation Evaluation of Coding Agents
- 2026SlopCodeBench: Benchmarking How Coding Agents Degrade Over Long-Horizon Iterative Tasks