Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A股研报摘要/卖方观点汇总/券商研报速览。当用户说"研报"、"券商研报"、"分析师怎么看"、"卖方观点"、"research digest"、"XX最新研报"、"目标价"、"评级"、"研究报告"、"机构怎么说"、"研报摘要"、"分析师观点"、"研报汇总"时触发。MUST USE when user asks about research report digest, analyst opinions/ratings/target prices, or sell-side consensus views on a stock. 通过 web 搜索汇总券商研报的核心观点、目标价、评级变化，帮助用户快速了解卖方共识。支持研报汇总风格（formal）和快速摘要风格（brief）。
.claude/skills/aifinlab-a-share-research-digest/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | 13% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-03 | ✓→✗ | ▼ Worse | -53% | 0% |
| case-04 | ✓→✗ | ▼ Worse | -47% | 0% |
| case-05 | ✓→✗ | ▼ Worse | -39% | 0% |
搜索关键词组合：
用于计算目标价上行/下行空间：
bashSCRIPTS="$SKILLS_ROOT/cn-stock-data/scripts" python "$SCRIPTS/cn_stock_data.py" quote --code [CODE]
## [股票名] 卖方观点汇总（截至 YYYY-MM）
### 一致预期概览
| 指标 | 数值 |
|------|------|
| 覆盖券商数 | N家 |
| 一致评级 | 买入/增持/中性 |
| 一致目标价（中位数） | ¥XX.XX |
| 当前价 | ¥XX.XX |
| 隐含上行空间 | +XX% |
### 各券商观点
| 券商 | 分析师 | 日期 | 评级 | 目标价 | 核心逻辑 |
|------|--------|------|------|--------|----------|
（按日期倒序）
### 多空分歧
- **看多逻辑**：...
- **看空/谨慎逻辑**：...
- **关键变量**：...（决定多空方向的核心因素）
### 风险提示
（汇总各券商提及的主要风险）[股票名] 研报速览：N家覆盖，一致评级[买入]，目标价中位数¥XX（当前¥XX，上行空间+XX%）
看多：[核心逻辑1-2句]
分歧：[关键分歧点]
风险：[主要风险]参见 references/research-digest-guide.md 中的评级体系对照表，统一映射为： 强烈推荐 → 买入 | 增持/推荐 → 增持 | 中性/持有 → 中性 | 减持/回避 → 减持
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 | fail→fail | 28,246 | 11,062 | -61% | 1 | 1 | 0% | 4,227 | 1,584 | -63% | 0 | 0 | — |
case-02 | fail→fail | 11,467 | 10,079 | -12% | 1 | 1 | 0% | 1,768 | 1,438 | -19% | 0 | 0 | — |
case-03 | pass→fail | 23,130 | 69,197 | +199% | 1 | 1 | 0% | 3,514 | 1,663 | -53% | 0 | 0 | — |
case-04 | pass→fail | 21,508 | 9,703 | -55% | 1 | 1 | 0% | 2,877 | 1,535 | -47% | 0 | 0 | — |
case-05 | pass→fail | 17,142 | 9,133 | -47% | 1 | 1 | 0% | 2,334 | 1,419 | -39% | 0 | 0 | — |
case-06 | pass→fail | 22,551 | 10,894 | -52% | 1 | 1 | 0% | 3,470 | 1,320 | -62% | 0 | 0 | — |
case-07 | fail→fail | 16,284 | 13,903 | -15% | 1 | 1 | 0% | 2,317 | 2,730 | +18% | 0 | 0 | — |
case-08 | fail→pass | 19,456 | 17,046 | -12% | 1 | 1 | 0% | 2,754 | 3,105 | +13% | 0 | 0 | — |
case-09 | fail→pass | 13,734 | 3,683 | -73% | 1 | 1 | 0% | 2,275 | 1,434 | -37% | 0 | 0 | — |
case-10 | pass→fail | 19,903 | 7,825 | -61% | 1 | 1 | 0% | 2,834 | 1,305 | -54% | 0 | 0 | — |
case-11 | pass→pass | 20,024 | 16,135 | -19% | 1 | 1 | 0% | 2,468 | 3,179 | +29% | 0 | 0 | — |
case-12 | fail→fail | 6,954 | 9,146 | +32% | 1 | 1 | 0% | 1,055 | 1,332 | +26% | 0 | 0 | — |
case-13 | fail→fail | 18,833 | 10,001 | -47% | 1 | 1 | 0% | 2,785 | 1,254 | -55% | 0 | 0 | — |
case-14 | pass→pass | 19,744 | 10,458 | -47% | 1 | 1 | 0% | 2,433 | 2,449 | +1% | 0 | 0 | — |
case-15 | fail→fail | 9,191 | 11,702 | +27% | 1 | 1 | 0% | 1,403 | 1,361 | -3% | 0 | 0 | — |
case-16 | pass→pass | 23,083 | 32,400 | +40% | 1 | 1 | 0% | 3,018 | 5,059 | +68% | 0 | 0 | — |
case-17 | pass→pass | 23,232 | 10,016 | -57% | 1 | 1 | 0% | 2,833 | 2,604 | -8% | 0 | 0 | — |
case-18 | pass→pass | 14,747 | 10,288 | -30% | 1 | 1 | 0% | 2,112 | 1,959 | -7% | 0 | 0 | — |
case-19 | pass→pass | 16,031 | 5,600 | -65% | 1 | 1 | 0% | 1,979 | 1,864 | -6% | 0 | 0 | — |
case-20 | fail→fail | 28,468 | 10,023 | -65% | 1 | 1 | 0% | 3,624 | 1,497 | -59% | 0 | 0 | — |
case-21 | fail→fail | 24,707 | 80,492 | +226% | 1 | 1 | 0% | 4,695 | 7,585 | +62% | 0 | 0 | — |
case-22 | pass→fail | 8,792 | 11,735 | +33% | 1 | 1 | 0% | 1,502 | 1,435 | -4% | 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, and 10 counted toward the lift figure. The other 12 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of -18 percentage points is the difference between those two pass rates over the 10 comparable cases. 8 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.