Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →梳理控股股东、实际控制人控制的其他企业，识别并解决同业竞争问题
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 34% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 52% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-20 | ✗→✓ | ▲ Improved | -15% | 0% |
本技能用于梳理控股股东、实际控制人控制的其他企业，识别是否存在同业竞争，并协助制定解决方案。
产品/服务相同或相似：
业务替代性：
| 核查对象 | 核查方式 | 核查内容 | |----------|----------|----------| | 控股股东 | 工商查询、访谈 | 控制的企业清单 | | 实际控制人 | 访谈、承诺函 | 投资情况、兼职情况 | | 董监高 | 调查问卷 | 对外投资、兼职 |
| 比对维度 | 发行人 | 竞争方 | 是否同业 | |----------|--------|--------|----------| | 主营业务 | | | | | 主要产品 | | | | | 目标客户 | | | | | 销售区域 | | | | | 核心技术 | | | |
收集控股股东、实际控制人控制的企业清单及业务信息
bashcd "C:\Users\51994\Desktop\skills\证券 2\10. 同业竞争梳理助手" python scripts/analyze_competition.py --input <关联方数据> --output <分析结果路径>
分析报告将包含：
根据分析结果制定并实施解决方案
markdown## 同业竞争分析报告 ### 关联方概况 - 控股股东控制企业：8 家 - 实际控制人控制企业：12 家 - 已排查企业：20 家 ### 同业竞争识别 | 竞争方 | 主营业务 | 与发行人关系 | 是否同业 | |--------|----------|--------------|----------| | A 公司 | XX 产品 | 控股股东控制 | 是 ⚠️ | | B 公司 | YY 服务 | 实际控制人控制 | 否 | | C 公司 | ZZ 产品 | 控股股东控制 | 相似 ⚠️ | ### 影响分析 - 竞争方 A 公司 2024 年收入：5,000 万元 - 占发行人同类业务收入比例：15% - 对发行人影响：中等 ### 解决方案 1. **A 公司同业竞争** - 方案：发行人收购 A 公司 100% 股权 - 进展：已签署收购协议 - 预计完成：2026-06-30 2. **C 公司业务相似** - 方案：C 公司变更经营范围，停止竞争业务 - 进展：已启动变更程序 - 预计完成：2026-05-31 ### 承诺措施 控股股东、实际控制人已出具《避免同业竞争承诺函》 ### 结论 同业竞争问题已制定解决方案，预计申报前完成整改。
scripts/analyze_competition.py - 同业竞争分析脚本references/competition_checklist.md - 同业竞争核查清单references/avoid_competition_commitment.md - 避免同业竞争承诺模板| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 28,262 | 28,537 | +1% | 1 | 1 | 0% | 4,223 | 4,190 | -1% | 0 | 0 | — |
case-02 | fail→fail | 22,925 | 22,206 | -3% | 1 | 1 | 0% | 3,437 | 4,762 | +39% | 0 | 0 | — |
case-03 | pass→fail | 28,443 | 26,207 | -8% | 1 | 1 | 0% | 4,227 | 5,190 | +23% | 0 | 0 | — |
case-04 | pass→pass | 24,355 | 25,164 | +3% | 1 | 1 | 0% | 3,719 | 5,241 | +41% | 0 | 0 | — |
case-05 | pass→pass | 24,447 | 24,827 | +2% | 1 | 1 | 0% | 3,933 | 5,397 | +37% | 0 | 0 | — |
case-06 | pass→pass | 22,355 | 25,425 | +14% | 1 | 1 | 0% | 3,593 | 5,403 | +50% | 0 | 0 | — |
case-07 | fail→pass | 16,048 | 6,620 | -59% | 1 | 1 | 0% | 2,508 | 2,378 | -5% | 0 | 0 | — |
case-08 | pass→pass | 20,033 | 19,943 | -0% | 1 | 1 | 0% | 3,025 | 4,164 | +38% | 0 | 0 | — |
case-09 | pass→pass | 20,585 | 20,903 | +2% | 1 | 1 | 0% | 3,054 | 4,450 | +46% | 0 | 0 | — |
case-10 | pass→pass | 20,879 | 21,065 | +1% | 1 | 1 | 0% | 3,056 | 4,408 | +44% | 0 | 0 | — |
case-11 | pass→pass | 17,750 | 19,670 | +11% | 1 | 1 | 0% | 2,650 | 4,145 | +56% | 0 | 0 | — |
case-12 | pass→pass | 13,893 | 13,462 | -3% | 1 | 1 | 0% | 1,961 | 3,213 | +64% | 0 | 0 | — |
case-13 | fail→pass | 18,375 | 16,115 | -12% | 1 | 1 | 0% | 2,985 | 4,000 | +34% | 0 | 0 | — |
case-14 | pass→pass | 19,172 | 18,098 | -6% | 1 | 1 | 0% | 2,866 | 3,981 | +39% | 0 | 0 | — |
case-15 | pass→pass | 21,175 | 23,911 | +13% | 1 | 1 | 0% | 3,023 | 4,534 | +50% | 0 | 0 | — |
case-16 | fail→pass | 19,270 | 21,374 | +11% | 1 | 1 | 0% | 2,875 | 4,363 | +52% | 0 | 0 | — |
case-17 | pass→pass | 20,167 | 19,467 | -3% | 1 | 1 | 0% | 2,883 | 3,989 | +38% | 0 | 0 | — |
case-18 | pass→pass | 21,134 | 19,462 | -8% | 1 | 1 | 0% | 3,115 | 4,227 | +36% | 0 | 0 | — |
case-19 | fail→pass | 17,811 | 6,720 | -62% | 1 | 1 | 0% | 2,570 | 2,261 | -12% | 0 | 0 | — |
case-20 | fail→pass | 15,131 | 4,852 | -68% | 1 | 1 | 0% | 2,219 | 1,884 | -15% | 0 | 0 | — |
case-21 | pass→pass | 20,810 | 16,818 | -19% | 1 | 1 | 0% | 3,157 | 3,878 | +23% | 0 | 0 | — |
case-22 | pass→pass | 19,536 | 21,367 | +9% | 1 | 1 | 0% | 2,826 | 4,196 | +48% | 0 | 0 | — |
case-23 | fail→pass | 7,660 | 4,721 | -38% | 1 | 1 | 0% | 1,216 | 2,084 | +71% | 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. 23 cases were attempted. The headline lift of +22 percentage points is the difference between those two pass rates over the 23 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
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.