{"slug": "agentic-harness-engineering-beta-requirement-centered-evolution-of-agent", "title": "Agentic Harness Engineering Beta: Requirement-Centered Evolution of Agent Harnesses", "summary": "A paper submitted to arXiv on 18 November 2026 introduces Agentic Harness Engineering Beta (AHE-Beta), a requirement-centered loop for evolving agent harnesses that raises pass@1 from 21.9% to 33.3% on held-out tasks while using 40.8% fewer tokens than Base Harness. AHE-Beta outperforms Claude Code, Meta-Harness and AHE on pass@1, and in a blind human A/B test real users preferred it in 31 of 48 valid pairs. Without further evolution, the gains transferred to GLM-5.3 and DeepSeek-V4.1-Flash with pass@1 improvements of 15.8% and 5.3% over Base Harness.", "body_md": "# Computer Science > Software Engineering\n\n  [Submitted on 18 Nov 2026]\n\n# Title:Agentic Harness Engineering Beta: Requirement-Centered Evolution of Agent Harnesses\n\n[View PDF](http://arxiv.org/pdf/2611.24924v1)\n\n[HTML (experimental)](https://arxiv.org/html/2611.24924v1)\n\nAbstract:Real-world harness engineering is driven by explicit and implicit user requirements, which guide development and validation. However, existing harness evolution methods often overlook implicit requirements and interaction needs, while using a single evolution agent to modify the harness and benchmark scores to select changes. This process often produces narrow patches and leads to benchmark-specific local optima. We introduce Agentic Harness Engineering Beta (AHE-Beta), a requirement-centered loop that brings real-world engineering practices into harness evolution. The loop consists of three steps: (1) User-driven evaluation lets simulated users raise their requirements as issues while they work with the harness over multiple turns. (2) Team-based modification implements those requirements as generalizable changes rather than task-specific patches. (3) Requirement-driven selection merges a change only when it resolves a verified requirement and passes review. Empirically, on held-out tasks, AHE-Beta raises pass@1 from 21.9% to 33.3% while using 40.8% fewer tokens than Base Harness, and achieves higher pass@1 than Claude Code, Meta-Harness, and AHE. In a blind human A/B test, real users prefer AHE-Beta in 31 of 48 valid pairs and rate it higher on all five dimensions, matching the directional preference of the user simulator. The human A/B results show that AHE-Beta achieves its goal of aligning harness evolution with real-world user requirements. Without further evolution, the gains transfer to GLM-5.3 and DeepSeek-V4.1-Flash with pass@1 improvements of 15.8% and 5.3% over Base Harness. Together, these results show that requirement-centered harness evolution is a more general method for improving agent capabilities.\n    \n\n### Additional Features\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer \n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers \n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps \n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations \n\n*(*[What are Smart Citations?](https://www.scite.ai/))\n# Code, Data and Media Associated with this Article\n\nalphaXiv \n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers \n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub \n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub \n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face \n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast \n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))\n# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower \n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender \n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))\n# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [**Learn more about arXivLabs**](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/agentic-harness-engineering-beta-requirement-centered-evolution-of-agent", "canonical_source": "http://arxiv.org/abs/2611.24924v1", "published_at": "2026-11-19 16:24:31+00:00", "updated_at": "2026-11-19 16:45:05.273568+00:00", "lang": "en", "topics": ["ai-agents", "ai-research", "large-language-models", "ai-tools"], "entities": ["Agentic Harness Engineering Beta", "AHE-Beta", "Claude Code", "Meta-Harness", "AHE", "GLM-5.3", "DeepSeek-V4.1-Flash", "arXiv"], "also_reported_by": [], "takeaway": "AHE-Beta's requirement-centered loop — simulated users filing issues, team-based modification, and merging only changes that resolve a verified requirement — lifted pass@1 from 21.9% to 33.3% on held-out tasks at 40.8% fewer tokens than Base Harness.", "recommendation": {"advice": "Evolve agent harnesses through simulated-user issues and team-based changes gated on verified requirements rather than single-agent benchmark-driven patches; the source measured pass@1 rising from 21.9% to 33.3% with 40.8% fewer tokens.", "playbook": "agent-architecture", "playbook_url": "https://wpnews.pro/playbooks/agent-architecture"}, "alternates": {"html": "https://wpnews.pro/news/agentic-harness-engineering-beta-requirement-centered-evolution-of-agent", "markdown": "https://wpnews.pro/news/agentic-harness-engineering-beta-requirement-centered-evolution-of-agent.md", "text": "https://wpnews.pro/news/agentic-harness-engineering-beta-requirement-centered-evolution-of-agent.txt", "jsonld": "https://wpnews.pro/news/agentic-harness-engineering-beta-requirement-centered-evolution-of-agent.jsonld"}}