[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85318-en":3,"doc-seo-85318-105":29,"detail-sidebar-cat-0-en-105":90},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},85318,1099514068365,"Aurelia","https://ap-avatar.wpscdn.com/avatar/10000253d8d9f28188e?_k=1776742907772140068",8,"Research & Report","Agentic Routing: The Harness-Native Data Flywheel","Large language model agents are executed through an orchestration harness that manages observation, context, control, action, state, and verification, while models themselves are increasingly specialized for coding, long-context recovery, tool use, reasoning, or low-latency responses. This shifts routing from a per-query cost-quality trick to a core systems optimization involving execution state, intermediate failures, and feedback loops. The work proposes harness-native agentic routing with step-level model selection or complementary ensembles, producing labeled execution records that train routers and models to improve cost-quality trade-offs and generate more traces under fixed budgets.","arXiv :2607 . 1 1399v 1 [ cs .CL] 13 Jul 2026  \nAgentic Routing: The Harness-Native Data Flywheel  \nTokenRhythm Technologies  \n[https://github.com/opensquilla/opensquilla](https://github.com/opensquilla/opensquilla)  \nAbstract  \nLarge language model agents are increasingly executed not by a single model call, but by an execution harness that manages observation, context, control, action, state, and verification. At the same time, frontier and open models are becoming structurally specialized: a model that is strong at code editing, long-context recovery, tool use, mathematical reasoning, or low-latency response may not dominate on the other axes. This makes model selection inside an agent a core systems problem rather than a per-query serving trick. Existing routing methods mostly optimize single-turn cost-quality trade-offs and therefore miss the execution state, intermediate failures, and feedback loops that make agents different from chat completion. We propose HARNESS-NATIVE agentic routing, a step-level routing paradigm that selects either a single best-fit model for cost-effective execution or multiple complementary models for ensemble-style accuracy improvement, conditioned on the full harness state. The key insight is that every routing decision naturally produces a structured data record—consisting of the query, harness state, model choice or model set, execution trace, outcome, and cost—whose labels are supplied by the environment rather than by the router itself. These records form a harness-native data flywheel: execution traces train better routers and harness-native models, which improve cost-quality trade-offs and generate more traces under the same budget. We instantiate this idea in OPENS QUILL A with a four-layer routing stack, an open LightGBM cold-start ranker, and a staged router-model path that turns logged arena records into progressively stronger routing policies. The report studies singleton and multi-model routing on agentic benchmarks including DRACO and PinchBench, and argues that agentic routing is not merely cost control, but a data engine for agent-native training.  \n1. Introduction  \nAn AI agent is traditionally defined by a simple decomposition: Agent = Model + Harness [29], where a single foundation model supplies the intelligence and the harness wraps it into a working system. As base models proliferate and become increasingly specialized—in coding, long-context recovery, reasoning, tool use, or low-latency response—no single model remains the best choice at every step, and the recent evolution of AI agents is better summarized by an updated decomposition: Agent = Base Models + Harness [10] . Base models [2, 26] provide latent capabilities such as language understanding, reasoning, code generation, tool-use priors, and multimodal perception, while the harness [6, 17, 27] provides the execution substrate that turns these capabilities into observable, controllable, and verifiable behavior. The two sides are  \nFigure 1 | Overview of the two operating regimes of agentic routing. The agent harness supplies the execution state to the router, which selects from a shared model pool. The user-selected mode determines the regime: in the cost-effective mode the router fields a single best-fit model, while in the high-accuracy mode it fields several complementary proposers whose outputs are fused by an aggregator. Verification and cost signals score each decision, and the resulting records feed back to improve the router.  \ncomplementary rather than hierarchical: a model without a harness remains a passive generator, and a harness without capable base models has no intelligence to orchestrate. Progress in agents is therefore an alternating process rather than a one-sided model-scaling story. Stronger base models make new behaviors possible; harness engineering converts those behaviors into reliable workflows; those workflows expose new bottlenecks such as context drift, brittle tool calls, poor recovery,","cbCairDql52d4Fcn","https://ap.wps.com/l/cbCairDql52d4Fcn","pdf",2358971,1,33,"English","en",105,"# Introduction\n## Agent = Base Models + Harness\n## Specialized models and the routing challenge\n## Step-level harness-native routing idea","[{\"question\":\"What problem does the proposed HARNESS-NATIVE agentic routing address?\",\"answer\":\"It addresses the mismatch between traditional routing methods that optimize only single-turn cost-quality trade-offs and the richer needs of agent execution, including harness state, intermediate failures, and feedback loops.\"},{\"question\":\"How does HARNESS-NATIVE routing make model choices during agent execution?\",\"answer\":\"It uses a step-level routing paradigm that selects either a single best-fit model for cost-effective execution or multiple complementary models for ensemble-style accuracy, conditioned on the full harness state.\"},{\"question\":\"What is the “harness-native data flywheel” in this work?\",\"answer\":\"Each routing decision generates structured records—query, harness state, model choice, execution trace, outcome, and cost—labeled by the environment; these records train better routers and harness-native models, improving cost-quality trade-offs and producing more traces within the same budget.\"}]",1784202456,83,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"agentic-routing-the-harness-native-data-flywheel","",{"@graph":35,"@context":84},[36,53,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/agentic-routing-the-harness-native-data-flywheel/85318/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What problem does the proposed HARNESS-NATIVE agentic routing address?","Question",{"text":74,"@type":75},"It addresses the mismatch between traditional routing methods that optimize only single-turn cost-quality trade-offs and the richer needs of agent execution, including harness state, intermediate failures, and feedback loops.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How does HARNESS-NATIVE routing make model choices during agent execution?",{"text":79,"@type":75},"It uses a step-level routing paradigm that selects either a single best-fit model for cost-effective execution or multiple complementary models for ensemble-style accuracy, conditioned on the full harness state.",{"name":81,"@type":72,"acceptedAnswer":82},"What is the “harness-native data flywheel” in this work?",{"text":83,"@type":75},"Each routing decision generates structured records—query, harness state, model choice, execution trace, outcome, and cost—labeled by the environment; these records train better routers and harness-native models, improving cost-quality trade-offs and producing more traces within the same budget.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":105,"slug":137},19,"General","general"]