HarnessBridge: Learnable Bidirectional Controller for LLM Agent Harness

Computer Science > Artificial Intelligence [Submitted on 11 Jun 2026] HarnessBridge: Learnable Bidirectional Controller for LLM Agent Harness Xiaoxuan Wang, Haixin Wang, Alexander Taylor, Jason Cong, Yizhou Sun, Wei Wang Large language models are increasingly deployed as agents for long-horizon tasks, yet their performance is shaped not only by model capability and environment design, but also by the harness that mediates agent--environment interaction. Existing harnesses are largely manually engineered, making them difficult to scale as trajectories grow longer and interactions become more complex. In this work, we ask whether harness can be generated by a learnable plug-in module that can be trained in an end-to-end fashion. We introduce HarnessBridge, a lightweight learnable harness controller that parameterizes the agent--environment interface as a bidirectional projection. HarnessBridge learns two bidirectional projections: observation projection, which distills raw trajectories into compact, decision-relevant states, and action projection, which converts proposed actions into executable transitions or trajectory-grounded rejections. We train HarnessBridge on a harness supervision dataset via unified instruction tuning. On Terminal-Bench~2.0 and SWE-bench Verified, HarnessBridge matches or surpasses strong specialized harnesses while substantially reducing token usage and trajectory length, and generalizes from smaller generators to larger commercial models. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2606.12882 [cs.AI]   (or arXiv:2606.12882v1 [cs.AI] for this version)   https://doi.org/10.48550/arXiv.2606.12882 Focus to learn more Submission history From: Xiaoxuan Wang [view email] [v1] Thu, 11 Jun 2026 04:18:37 UTC (1,127 KB) Access Paper: view license Current browse context: cs.AI < prev   |   next > new | recent | 2026-06 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)