Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →公告解读助手 - 再融资版。专注于再融资公告解读，分析定增/配股/可转债方案、融资规模、发行价格、募投项目及对股东权益的影响。 **触发场景**： - 用户粘贴再融资公告（定增/配股/可转债），需要解读 - 用户问"这个定增怎么样"、"融资规模大吗" - 需要分析融资方案、发行价格、募投项目 - 需要判断再融资对股东权益的影响 **关键词**："定增"、"配股"、"可转债"、"再融资"、"融资"、"发行股份"、"募投项目"、"稀释"
.claude/skills/aifinlab-announcement-interpretation-refinancing/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-14 | ✗→✓ | ▲ Improved | 42% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 70% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 22% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-20 | ✗→✓ | ▲ Improved | -26% | 0% |
你是一名经验丰富的分析师，擅长解读再融资公告，帮助投资者判断融资方案质量和对股东权益的影响。
| 类型 | 定义 | 股权稀释 | 资金压力 | 利好程度 | |------|------|----------|----------|----------| | 定增 | 向特定对象发行股份 | 有稀释 | 无还款压力 | 中等 | | 配股 | 向原股东按比例配售 | 不配则稀释 | 无还款压力 | 偏利空 | | 可转债 | 发行可转换债券 | 转股后稀释 | 有利息压力 | 中等 | | 公开增发 | 向公众发行股份 | 有稀释 | 无还款压力 | 偏利空 |
【XX 公司再融资公告解读】
融资类型：[定增/配股/可转债]
融资规模：XX 亿，占总市值 XX%
发行价格：XX 元，较现价折价 XX%
股权稀释：预计稀释 XX%
募投项目：[项目 1]、[项目 2]
投资判断：[利好/利空程度 + 简要理由]# 【XX 公司再融资公告解读】
## 融资方案概览
| 项目 | 内容 |
|------|------|
| 融资类型 | |
| 融资规模 | |
| 发行价格 | |
| 发行对象 | |
| 锁定期 | |
| 募投项目 | |
## 融资规模分析
- 绝对金额：XX 亿元
- 占总市值：XX%
- 规模判断：[大/中/小]
## 发行价格分析
- 发行价：XX 元
- 较现价折价：XX%
- 折价判断：[高/中/低]
## 股权稀释分析
- 发行前股本：XX 亿股
- 发行后股本：XX 亿股
- 稀释比例：XX%
- EPS 摊薄：预计 XX%
## 募投项目分析
### 项目 1
- 投资金额：XX 亿
- 项目内容：[描述]
- 预期回报：[分析]
### 项目 2
- [同上]
## 认购对象分析（定增）
- 认购方：[大股东/战投/机构]
- 信心信号：[分析]
## 投资含义
- 短期影响：[股价反应预判]
- 中期影响：[EPS 变化/业务发展]
## 风险提示
- 融资失败风险
- 募投项目不及预期
- 股权稀释风险输出前自查：
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-08 | pass→pass | 12,143 | 10,576 | -13% | 1 | 1 | 0% | 2,131 | 3,142 | +47% | 0 | 0 | — |
case-01 | fail→fail | 28,041 | 27,648 | -1% | 1 | 1 | 0% | 4,675 | 5,102 | +9% | 0 | 0 | — |
case-02 | fail→fail | 24,402 | 21,635 | -11% | 1 | 1 | 0% | 4,081 | 5,107 | +25% | 0 | 0 | — |
case-03 | pass→pass | 21,651 | 20,650 | -5% | 1 | 1 | 0% | 2,791 | 3,864 | +38% | 0 | 0 | — |
case-04 | fail→fail | 17,922 | 20,399 | +14% | 1 | 1 | 0% | 2,688 | 3,832 | +43% | 0 | 0 | — |
case-05 | pass→pass | 21,189 | 23,585 | +11% | 1 | 1 | 0% | 3,154 | 4,310 | +37% | 0 | 0 | — |
case-06 | pass→pass | 19,972 | 13,977 | -30% | 1 | 1 | 0% | 3,501 | 3,790 | +8% | 0 | 0 | — |
case-07 | pass→pass | 19,597 | 18,039 | -8% | 1 | 1 | 0% | 3,181 | 4,283 | +35% | 0 | 0 | — |
case-09 | pass→pass | 22,727 | 21,255 | -6% | 1 | 1 | 0% | 2,931 | 4,220 | +44% | 0 | 0 | — |
case-10 | pass→pass | 22,291 | 36,391 | +63% | 1 | 1 | 0% | 3,404 | 4,729 | +39% | 0 | 0 | — |
case-11 | pass→pass | 25,441 | 20,789 | -18% | 1 | 1 | 0% | 3,281 | 3,752 | +14% | 0 | 0 | — |
case-12 | pass→pass | 23,979 | 18,482 | -23% | 1 | 1 | 0% | 2,823 | 3,864 | +37% | 0 | 0 | — |
case-13 | pass→pass | 22,609 | 19,713 | -13% | 1 | 1 | 0% | 2,741 | 4,150 | +51% | 0 | 0 | — |
case-14 | fail→pass | 22,724 | 18,380 | -19% | 1 | 1 | 0% | 2,781 | 3,939 | +42% | 0 | 0 | — |
case-15 | fail→pass | 13,869 | 15,793 | +14% | 1 | 1 | 0% | 2,110 | 3,581 | +70% | 0 | 0 | — |
case-16 | fail→pass | 22,001 | 7,493 | -66% | 1 | 1 | 0% | 2,038 | 2,493 | +22% | 0 | 0 | — |
case-17 | fail→fail | 25,831 | 27,224 | +5% | 1 | 1 | 0% | 3,934 | 4,951 | +26% | 0 | 0 | — |
case-18 | pass→pass | 23,138 | 17,234 | -26% | 1 | 1 | 0% | 3,364 | 3,838 | +14% | 0 | 0 | — |
case-19 | fail→pass | 15,911 | 8,043 | -49% | 1 | 1 | 0% | 2,442 | 2,360 | -3% | 0 | 0 | — |
case-20 | fail→pass | 17,881 | 5,335 | -70% | 1 | 1 | 0% | 2,541 | 1,868 | -26% | 0 | 0 | — |
case-21 | pass→pass | 12,668 | 6,174 | -51% | 1 | 1 | 0% | 2,058 | 2,303 | +12% | 0 | 0 | — |
case-22 | pass→pass | 13,046 | 7,975 | -39% | 1 | 1 | 0% | 1,909 | 2,599 | +36% | 0 | 0 | — |
case-23 | pass→pass | 15,984 | 16,577 | +4% | 1 | 1 | 0% | 2,352 | 3,329 | +42% | 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.
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.