Agentic AI systems act through tools and sub-agents, yet the controls meant to bound their financial and environmental cost still sit on dashboards evaluated beside or after execution. Green SARC applies the SARC governance-by-architecture framework -- four enforcement sites in the agent loop -- to FinOps and GreenOps, contributing the theory of what to enforce and how to predict it. We report four policy-independent results. (i) The unconstrained "State Snowball" is $Θ(n^2)$ in loop depth; on 3,000 real multi-step plans (SWE-rebench) it holds on 100%, with median curvature $\hat{c}_2=216$ exceeding the linear-accretion prediction $p/2=134$ -- real plans accrete faster than the model. (ii) On real residuals the Normal-$σ$ gate under-covers (92% at nominal 95%); split-conformal calibration holds (95.2%). (iii) A soft Lagrangian penalty tuned to the budget in expectation breaches it on 91.5% of seeds; the architectural gate breaches 0%. (iv) Under binding budgets the gate's over-budget incidence is 0% on synthetic and real (BurstGPT) arrivals. End-to-end token/USD/carbon savings (47--55%) are real but policy-dependent in magnitude -- set by a scope-cap knob, not by gate rejections. The library is open-source, dependency-free, and ships a regeneration script for every cited number.

翻译：自主AI系统通过工具和子智能体执行操作，但旨在约束其财务与环境成本的控制措施仍停留在运行中或运行后通过仪表盘进行评估的层面。Green SARC将SARC架构治理框架（智能体循环中的四个强制执行位点）应用于FinOps和GreenOps，提出了需要强制执行什么以及如何预测其影响的理论。我们报告了四项与具体策略无关的结论：（i）无约束的“状态雪球”在循环深度上具有Θ(n²)复杂度；在3,000个真实多步计划（SWE-rebench）中，该结论100%成立，中位曲率̂c₂=216超过线性累积预测值p/2=134——真实计划的累积速度比模型更快。（ii）在实际残差上，正态-σ门控在名义置信水平95%下的覆盖率为92%，存在覆盖不足；分裂共形校准后覆盖率提升至95.2%。（iii）根据预算期望值调优的软拉格朗日惩罚项在91.5%的随机种子实验中违反预算约束；而架构门控的违反率为0%。（iv）在严格预算约束下，门控在合成数据和真实数据（BurstGPT）到达序列上的超预算发生率均为0%。端到端的Token/美元/碳节省率（47–55%）真实存在，但其幅度取决于具体策略——由范围-能力控制旋钮设定，而非门控拒绝机制。该库为开源且无依赖项，并附有对每个引用数据进行再生成的脚本。