发表机构
G.S. s.r.o; PwC Austria（G.S.有限责任公司; 普华永道奥地利公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文针对大型企业知识工作的代码维护痛点，提出基于 harness 范式的架构，通过统一核心、优化机制缩小研究发现与企业采用的治理差距。
AI 中文摘要
前沿大模型已大幅降低自定义代码编写的成本：领域专家处理小众问题仅需一个下午，但代码的审查与维护成本并未随之降低，各解决方案间存在差异，理解某一方案需从头阅读其代码库。大型企业通常采用集中管控的方案：最差是现成产品，较好是针对用例定制的图编排框架，或是用作编排器的低代码平台，这些方案每次都需定制且范围有限。企业未考虑第三种能摆脱上述两种限制的方案：harness 范式。近期研究将编码智能体 harness 视为企业基础设施而非编码工具，得出三项核心结论：harness 在任务层面已足够，在企业工作中表现优于更复杂的架构（arXiv:2604.00073、arXiv:2604.13107）；harness 的选择对智能体基准测试结果的方差影响最大，超过模型选择的影响（arXiv:2605.23950）；这一发现与企业采用之间的差距在于治理问题（arXiv:2605.10223、arXiv:2605.18747）。本文提出一种缩小该差距的架构：一个 harness 无需修改即可作为核心运行，所有部署间代码保持一致，因此审查生成内容仅需阅读其指令文件。第4节介绍四种机制：凭证范围工具，每个后端配备一个通用请求工具和范围凭证，而非手动构建的方法；授权逻辑位于 harness 外部，使同一工件可作为 cron 核心、聊天界面引擎和终端工具运行；注册是代码推送的副作用，将审计缩减为对文本文件的审查。该架构基于 microcc（<此链接>）构建，是本文的参考 harness。
英文摘要
Frontier models have collapsed the cost of writing custom code: a niche problem a specialist sees in their own domain now costs an afternoon. The cost of reviewing and maintaining that code hasn't collapsed. Each solution drifts from the next; understanding one means reading its codebase from scratch. Large enterprises build something centrally governed instead: at worst an off-the-shelf product, at best a graph-orchestration framework wired bespoke per use case, or a low-code platform used as the orchestrator. These are custom every time and limited in scope. Enterprises don't weigh a third option that escapes both constraints: the harness paradigm. Recent work treats the coding-agent harness as enterprise infrastructure rather than a coding tool, converging on three findings: harnesses suffice at the task level and outperform more elaborate architectures on enterprise work (arXiv:2604.00073, arXiv:2604.13107); harness choice accounts for most of the variance in agent benchmark results, more than model choice does (arXiv:2605.23950); and the gap between that finding and enterprise adoption is governance (arXiv:2605.10223, arXiv:2605.18747). We propose an architecture that closes that gap. One harness runs unmodified as the backbone; the code stays identical across every deployment, so reviewing what gets built collapses to reading its instructions file. Section 4 gives four mechanisms: credential-scoped tooling, where each backend gets one generic request tool and a scoped credential instead of a hand-built method; authorization logic outside the harness, so one artifact runs as a cron backbone, a chat-surface engine, and a terminal tool; registration is a side effect of pushing code, collapsing an audit a review of a text file. Built on microcc (<https://pypi.org/project/micro-cc/>), our reference harness.
Comments21 pages, 3 figures
