Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →公告解读助手 - 中标公告版。专注于项目中标/投标公告解读，分析中标金额、项目类型、招标方资质、项目周期及对业绩的影响。 **触发场景**： - 用户粘贴中标公告/投标公告，需要解读 - 用户问"这个中标怎么样"、"金额大吗" - 需要分析中标金额、项目类型、招标方 - 需要判断中标对业绩的影响 **关键词**："中标"、"投标"、"中标公告"、"项目中标"、"招标"、"标段"、"中标金额"、"预中标"
.claude/skills/aifinlab-announcement-interpretation-bid-win/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 30% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 73% | 0% |
你是一名经验丰富的分析师，擅长解读中标公告，帮助投资者判断中标项目质量和业绩贡献。
| 要素 | 关注点 | 利好程度 | |------|--------|----------| | 中标金额 | 金额大小、占营收比例 | 比例越大越利好 | | 项目类型 | 工程/产品/服务 | 核心业务利好 | | 招标方 | 政府/国企/民企 | 政府/国企更可靠 | | 项目周期 | 履行时间 | 周期短见效快 | | 中标方式 | 公开招标/邀请招标 | 公开更透明 |
【XX 公司中标公告解读】
中标金额：XX 亿，占上年度营收 XX%
项目类型：[工程/产品/服务]
招标方：[名称 + 资质]
项目周期：XX 个月/年
业绩影响：预计贡献营收 XX 亿，利润 XX 亿
投资判断：[利好程度 + 简要理由]# 【XX 公司中标公告解读】
## 中标概览
| 项目 | 内容 |
|------|------|
| 中标金额 | |
| 项目名称 | |
| 项目类型 | |
| 招标方 | |
| 招标方资质 | |
| 项目周期 | |
| 中标方式 | |
## 中标金额分析
- 绝对金额：XX 亿元
- 占上年度营收：XX%
- 重要性：[重大/中等/小]
## 项目类型分析
- 类型：[工程/产品/服务]
- 是否核心业务：[是/否]
- 战略意义：[分析]
## 招标方分析
- 招标方：[名称]
- 资质：[政府/国企/民企]
- 付款能力：[分析]
## 项目周期分析
- 履行期限：XX 个月
- 收入确认：[按进度/一次性]
- 业绩贡献：预计 XX 年确认 XX 亿
## 盈利分析
- 预计毛利率：XX%（如有）
- 对比公司平均：[高/低]XXpct
- 预计利润贡献：XX 亿
## 历史中标对比
- 历史中标记录：[如有]
- 中标率：[如有]
## 投资含义
- 短期影响：[股价反应预判]
- 中期影响：[业绩增厚]
## 风险提示
- 项目履约风险
- 招标方付款风险
- 毛利率不及预期输出前自查：
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-09 | pass→pass | 18,623 | 21,439 | +15% | 1 | 1 | 0% | 2,819 | 3,935 | +40% | 0 | 0 | — |
case-01 | fail→fail | 21,977 | 24,088 | +10% | 1 | 1 | 0% | 3,482 | 3,943 | +13% | 0 | 0 | — |
case-02 | fail→fail | 32,053 | 27,929 | -13% | 1 | 1 | 0% | 4,155 | 5,893 | +42% | 0 | 0 | — |
case-03 | fail→fail | 10,674 | 16,691 | +56% | 1 | 1 | 0% | 1,754 | 2,392 | +36% | 0 | 0 | — |
case-04 | fail→pass | 18,117 | 14,256 | -21% | 1 | 1 | 0% | 2,689 | 3,484 | +30% | 0 | 0 | — |
case-05 | fail→fail | 11,699 | 15,427 | +32% | 1 | 1 | 0% | 1,888 | 3,424 | +81% | 0 | 0 | — |
case-06 | pass→pass | 15,476 | 17,399 | +12% | 1 | 1 | 0% | 2,281 | 3,540 | +55% | 0 | 0 | — |
case-07 | pass→pass | 18,276 | 16,673 | -9% | 1 | 1 | 0% | 2,878 | 3,252 | +13% | 0 | 0 | — |
case-08 | pass→pass | 21,517 | 15,845 | -26% | 1 | 1 | 0% | 2,869 | 3,479 | +21% | 0 | 0 | — |
case-10 | pass→pass | 14,790 | 22,384 | +51% | 1 | 1 | 0% | 2,084 | 3,992 | +92% | 0 | 0 | — |
case-11 | fail→pass | 26,078 | 24,908 | -4% | 1 | 1 | 0% | 3,936 | 3,345 | -15% | 0 | 0 | — |
case-12 | pass→pass | 16,559 | 19,276 | +16% | 1 | 1 | 0% | 2,972 | 3,926 | +32% | 0 | 0 | — |
case-13 | pass→pass | 18,448 | 16,326 | -12% | 1 | 1 | 0% | 1,887 | 3,278 | +74% | 0 | 0 | — |
case-14 | pass→pass | 42,109 | 19,345 | -54% | 1 | 1 | 0% | 3,245 | 3,916 | +21% | 0 | 0 | — |
case-15 | fail→fail | 8,320 | 8,718 | +5% | 1 | 1 | 0% | 1,320 | 2,374 | +80% | 0 | 0 | — |
case-16 | fail→pass | 25,437 | 14,470 | -43% | 1 | 1 | 0% | 3,687 | 3,779 | +2% | 0 | 0 | — |
case-17 | fail→pass | 20,046 | 14,476 | -28% | 1 | 1 | 0% | 2,649 | 2,805 | +6% | 0 | 0 | — |
case-18 | pass→pass | 18,914 | 17,085 | -10% | 1 | 1 | 0% | 2,548 | 3,951 | +55% | 0 | 0 | — |
case-19 | fail→pass | 11,958 | 6,856 | -43% | 1 | 1 | 0% | 1,333 | 2,302 | +73% | 0 | 0 | — |
case-20 | fail→pass | 15,304 | 16,087 | +5% | 1 | 1 | 0% | 2,190 | 3,345 | +53% | 0 | 0 | — |
case-21 | fail→fail | 20,110 | 18,830 | -6% | 1 | 1 | 0% | 2,918 | 4,046 | +39% | 0 | 0 | — |
case-22 | fail→fail | 18,568 | 14,966 | -19% | 1 | 1 | 0% | 2,374 | 3,500 | +47% | 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 +27 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.