Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →公告解读助手 - 并购版。专注于并购重组公告解读，分析交易方案、标的资产、交易对价、业绩承诺及对上市公司的影响。 **触发场景**： - 用户粘贴并购重组公告，需要解读 - 用户问"这个并购怎么样"、"收购标的好吗" - 需要分析交易方案、标的资产质量、交易对价 - 需要判断并购对上市公司的影响 **关键词**："并购"、"重组"、"收购"、"资产注入"、"交易对价"、"业绩承诺"、"发行股份"、"现金收购"
.claude/skills/aifinlab-announcement-interpretation-ma/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✗→✓ | ▲ Improved | 43% | 0% |
| case-04 | ✓→✗ | ▼ Worse | 33% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | 28% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -4% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 44% | 0% |
你是一名经验丰富的分析师，擅长解读并购重组公告，帮助投资者判断交易质量和投资含义。
| 要素 | 关注点 | 利好程度 | |------|--------|----------| | 标的资产 | 资产质量、盈利能力 | 优质资产利好 | | 交易对价 | 估值水平、支付方式 | 合理对价利好 | | 业绩承诺 | 承诺金额、完成率 | 有承诺更可靠 | | 协同效应 | 业务协同、成本协同 | 协同明显利好 | | 融资方式 | 现金/股份/配套融资 | 现金>股份 |
【XX 公司并购公告解读】
交易方案：收购 [标的名称]XX% 股权，交易对价 XX 亿
标的资产：[业务 + 盈利能力]
估值水平：PE XX 倍，较同行业 [溢价/折价]XX%
业绩承诺：未来 3 年承诺净利 XX/XX/XX 亿
协同效应：[业务/成本协同]
投资判断：[利好程度 + 简要理由]# 【XX 公司并购公告解读】
## 交易方案概览
| 项目 | 内容 |
|------|------|
| 标的资产 | |
| 收购比例 | |
| 交易对价 | |
| 支付方式 | |
| 业绩承诺 | |
| 交易后股权结构 | |
## 标的资产分析
### 业务概况
- 主营业务：[描述]
- 行业地位：[分析]
### 财务数据
| 指标 | 2022 | 2023 | 2024Q1 |
|------|------|------|--------|
| 营收 (亿) | | | |
| 净利 (亿) | | | |
| 毛利率 (%) | | | |
## 交易对价分析
- 交易金额：XX 亿元
- 估值水平：PE XX 倍、PB XX 倍
- 对比同行业：[分析]
- 溢价率：较净资产溢价 XX%
## 业绩承诺分析
- 承诺金额：2024-2026 年 XX/XX/XX 亿
- 承诺增速：CAGR XX%
- 补偿条款：[分析]
## 协同效应分析
- 业务协同：[分析]
- 成本协同：[分析]
## 投资含义
- 短期影响：[股价反应预判]
- 中期影响：[EPS 增厚/业务拓展]
## 风险提示
- 整合不及预期
- 业绩承诺未完成
- 审批风险输出前自查：
python# 调用 skill result = run_skill({ "param1": "value1", "param2": "value2" })
bashpython scripts/run_skill.py --input data.json
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 14,865 | 16,895 | +14% | 1 | 1 | 0% | 2,054 | 2,619 | +28% | 0 | 0 | — |
case-02 | pass→pass | 30,704 | 21,256 | -31% | 1 | 1 | 0% | 4,998 | 4,775 | -4% | 0 | 0 | — |
case-03 | fail→fail | 25,847 | 26,482 | +2% | 1 | 1 | 0% | 3,725 | 5,021 | +35% | 0 | 0 | — |
case-04 | pass→fail | 20,922 | 23,752 | +14% | 1 | 1 | 0% | 3,075 | 4,077 | +33% | 0 | 0 | — |
case-05 | pass→pass | 15,179 | 13,867 | -9% | 1 | 1 | 0% | 2,106 | 3,042 | +44% | 0 | 0 | — |
case-06 | pass→pass | 20,617 | 21,227 | +3% | 1 | 1 | 0% | 3,045 | 4,348 | +43% | 0 | 0 | — |
case-07 | pass→pass | 25,748 | 17,711 | -31% | 1 | 1 | 0% | 3,318 | 3,928 | +18% | 0 | 0 | — |
case-08 | pass→pass | 21,057 | 18,381 | -13% | 1 | 1 | 0% | 3,112 | 4,014 | +29% | 0 | 0 | — |
case-09 | fail→fail | 16,016 | 13,255 | -17% | 1 | 1 | 0% | 2,281 | 2,815 | +23% | 0 | 0 | — |
case-10 | fail→pass | 20,022 | 39,418 | +97% | 1 | 1 | 0% | 2,525 | 3,616 | +43% | 0 | 0 | — |
case-11 | pass→pass | 15,393 | 11,723 | -24% | 1 | 1 | 0% | 1,955 | 2,856 | +46% | 0 | 0 | — |
case-12 | fail→fail | 24,430 | 22,665 | -7% | 1 | 1 | 0% | 3,054 | 4,208 | +38% | 0 | 0 | — |
case-13 | fail→fail | 15,647 | 17,526 | +12% | 1 | 1 | 0% | 2,698 | 3,652 | +35% | 0 | 0 | — |
case-14 | pass→pass | 22,782 | 17,357 | -24% | 1 | 1 | 0% | 2,678 | 3,577 | +34% | 0 | 0 | — |
case-15 | pass→pass | 16,379 | 9,936 | -39% | 1 | 1 | 0% | 2,196 | 2,728 | +24% | 0 | 0 | — |
case-16 | pass→pass | 21,448 | 17,381 | -19% | 1 | 1 | 0% | 3,157 | 3,818 | +21% | 0 | 0 | — |
case-17 | pass→pass | 21,498 | 13,738 | -36% | 1 | 1 | 0% | 3,192 | 3,363 | +5% | 0 | 0 | — |
case-18 | pass→pass | 18,672 | 23,901 | +28% | 1 | 1 | 0% | 2,836 | 4,107 | +45% | 0 | 0 | — |
case-19 | pass→pass | 19,489 | 24,746 | +27% | 1 | 1 | 0% | 3,606 | 5,009 | +39% | 0 | 0 | — |
case-20 | pass→pass | 28,335 | 25,074 | -12% | 1 | 1 | 0% | 3,674 | 4,862 | +32% | 0 | 0 | — |
case-21 | pass→pass | 25,146 | 17,090 | -32% | 1 | 1 | 0% | 3,450 | 3,889 | +13% | 0 | 0 | — |
case-22 | fail→fail | 24,872 | 20,986 | -16% | 1 | 1 | 0% | 3,106 | 3,890 | +25% | 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 0 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is 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.