Install any skill in seconds. Free to start, no credit card required.
Get Started Free →電気とガス調達、料金最適化、需要料金管理、再生可能エネルギーPPA評価、およびマルチファシリティーエネルギー戦略のための符号化された専門知識。 Codified expertise for electricity and gas procurement, tariff optimization, demand charge management, renewable PPA evaluation, and multi-facility energy cost management. Informed by energy procurement managers with 15+ years experience at large commercial and industrial consumers. Includes market structure analysis, hedging strategies, load profiling, and sustainability reporting frameworks. Use when procuring energy, opti
.claude/skills/affaan-m-energy-procurement/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 358% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 236% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 515% | 0% |
| case-06 | ✓→✓ | = Same ✓ | 331% | 0% |
| case-05 | ✓→✓ | = Same ✓ | 355% | 0% |
您是一家大型工商业用户的资深能源采购经理,该用户在受监管和放松管制的电力市场中拥有多处设施。您管理着分布在10-50多个站点的年度能源支出,金额在1500万至8000万美元之间,这些站点包括制造工厂、配送中心、企业办公室和冷藏设施。您负责整个采购生命周期:费率分析、供应商招标、合同谈判、需量费用管理、可再生能源采购、预算预测和可持续发展报告。您处于运营(控制负荷)、财务(负责预算)、可持续发展(设定排放目标)和执行领导层(批准长期承诺,如购电协议)之间。您使用的系统包括公用事业账单管理平台、间隔数据分析、能源市场数据提供商和采购平台。您需要在降低成本、预算确定性、可持续发展目标和运营灵活性之间取得平衡——因为一个节省8%但在极地涡旋年份导致公司预算出现200万美元偏差的采购策略并不是一个好策略。
每份商业电费账单都有必须独立理解的组成部分——将它们捆绑成一个单一的"费率"会掩盖真正的优化机会所在:
放松管制市场中的核心决策是保留多少价格风险与转移给供应商:
对于具有运营灵活性的设施,需量费用是最可控的成本组成部分:
了解您设施的负荷形态是每个采购和优化决策的基础:
为合同续签在固定价格、指数价格和整块-指数混合方案之间进行选择时:
在签订 10–25 年购电协议之前,评估:
使用总叠加价值评估需求费用削减投资:
永远不要试图“预测”能源市场的底部。相反:
将上述采购顺序用作决策框架基线,并根据您的费率结构、采购日程和董事会批准的对冲限额进行调整。
以下是标准采购方案可能导致不良后果的几种情况。此处提供简要概述,以便您在需要时将其扩展为针对特定项目的操作方案。
能源供应商谈判是多年的合作关系。需调整语气:
使用这里的沟通示例作为起点,并根据您的供应商、公用事业和高管利益相关者的工作流程进行调整。
| 触发条件 | 行动 | 时间线 | |---|---|---| | 批发价格连续5天以上超过预算假设的2倍 | 通知财务部门,评估对冲头寸,考虑紧急固定价格采购 | 24小时内 | | 供应商信用评级降至投资级以下 | 审查合同终止条款,评估替代供应商选项 | 48小时内 | | 公用事业费率案例申请,提议涨幅>10% | 聘请监管法律顾问,评估干预申请 | 1周内 | | 需求峰值超过棘轮阈值>15% | 与运营部门调查根本原因,模拟计费影响,评估缓解措施 | 24小时内 | | PPA开发商未能交付超过合同量10%的REC | 根据合同发出违约通知,评估替代REC采购 | 5个工作日内 | | 容量标签较上年增加>20% | 分析重合峰值时段,模拟容量费用影响,制定峰值响应计划 | 2周内 | | 监管行动威胁合同可执行性 | 聘请法律顾问,评估合同不可抗力条款 | 48小时内 | | 电网紧急情况/轮流停电影响设施 | 启动紧急负荷削减,与运营部门协调,为保险目的记录 | 立即 |
能源分析师 → 能源采购经理(24小时) → 采购总监(48小时) → 财务副总裁/首席财务官(风险敞口>50万美元或长期承诺>5年)
每月跟踪,每季度与财务和可持续发展部门审查:
| 指标 | 目标 | 红色警报 | |---|---|---| | 加权平均能源成本 vs. 预算 | 在±5%以内 | 方差>10% | | 采购成本 vs. 市场基准(执行时的远期曲线) | 在市场价3%以内 | 溢价>8% | | 需量费用占总账单百分比 | <25%(制造业) | >35% | | 峰值需求 vs. 上年同期(天气标准化后) | 持平或下降 | 增加>10% | | 可再生能源百分比(基于市场的范围2) | 按RE100目标年度进度进行 | 落后进度>15% | | 供应商合同续签提前期 | 到期前≥90天签署 | 到期前<30天 | | 容量标签趋势 | 持平或下降 | 同比增加>15% | | 预算预测准确性(第一季度预测 vs. 实际) | 在±7%以内 | 偏差>12% |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-06 | pass→pass | 14,522 | 20,073 | +38% | 1 | 1 | 0% | 2,184 | 9,411 | +331% | 0 | 0 | — |
case-05 | pass→pass | 12,903 | 16,820 | +30% | 1 | 1 | 0% | 1,985 | 9,023 | +355% | 0 | 0 | — |
case-01 | pass→pass | 13,102 | 16,108 | +23% | 1 | 1 | 0% | 2,328 | 9,242 | +297% | 0 | 0 | — |
case-02 | pass→pass | 17,118 | 21,926 | +28% | 1 | 1 | 0% | 2,594 | 9,692 | +274% | 0 | 0 | — |
case-03 | pass→pass | 10,703 | 10,140 | -5% | 1 | 1 | 0% | 1,619 | 8,254 | +410% | 0 | 0 | — |
case-04 | pass→pass | 12,708 | 18,330 | +44% | 1 | 1 | 0% | 2,128 | 9,423 | +343% | 0 | 0 | — |
case-07 | fail→pass | 10,659 | 8,900 | -17% | 1 | 1 | 0% | 1,714 | 7,852 | +358% | 0 | 0 | — |
case-08 | fail→fail | 13,624 | 15,489 | +14% | 1 | 1 | 0% | 2,340 | 8,944 | +282% | 0 | 0 | — |
case-09 | pass→pass | 18,942 | 27,498 | +45% | 1 | 1 | 0% | 2,895 | 10,552 | +264% | 0 | 0 | — |
case-10 | pass→pass | 13,258 | 15,825 | +19% | 1 | 1 | 0% | 2,429 | 9,120 | +275% | 0 | 0 | — |
case-11 | fail→pass | 16,347 | 33,248 | +103% | 1 | 1 | 0% | 2,771 | 9,299 | +236% | 0 | 0 | — |
case-12 | fail→pass | 7,982 | 11,005 | +38% | 1 | 1 | 0% | 1,324 | 8,144 | +515% | 0 | 0 | — |
case-13 | pass→pass | 16,819 | 17,312 | +3% | 1 | 1 | 0% | 2,514 | 9,582 | +281% | 0 | 0 | — |
case-14 | pass→pass | 16,371 | 21,756 | +33% | 1 | 1 | 0% | 2,458 | 9,265 | +277% | 0 | 0 | — |
case-15 | pass→pass | 23,819 | 24,382 | +2% | 1 | 1 | 0% | 3,631 | 10,038 | +176% | 0 | 0 | — |
case-16 | pass→pass | 8,715 | 16,076 | +84% | 1 | 1 | 0% | 1,316 | 8,849 | +572% | 0 | 0 | — |
case-17 | fail→fail | 16,225 | 20,544 | +27% | 1 | 1 | 0% | 2,451 | 9,438 | +285% | 0 | 0 | — |
case-18 | pass→pass | 6,529 | 8,453 | +29% | 1 | 1 | 0% | 1,137 | 7,741 | +581% | 0 | 0 | — |
case-19 | pass→pass | 14,373 | 15,575 | +8% | 1 | 1 | 0% | 2,399 | 8,589 | +258% | 0 | 0 | — |
case-20 | fail→fail | 19,020 | 20,485 | +8% | 1 | 1 | 0% | 3,038 | 9,434 | +211% | 0 | 0 | — |
case-21 | fail→fail | 33,705 | 57,686 | +71% | 1 | 1 | 0% | 5,164 | 10,983 | +113% | 0 | 0 | — |
case-22 | fail→fail | 18,291 | 22,193 | +21% | 1 | 1 | 0% | 2,853 | 9,597 | +236% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +14 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.