Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →财经新闻与公告Skill - 提供热点财经新闻、公司公告、券商研报、情感分析 via AkShare/东方财富
.claude/skills/aifinlab-akshare-news/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 156% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 156% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 182% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 159% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 28% | 0% |
| 属性 | 内容 | |:---|:---| | 名称 | akshare-news | | 版本 | 1.1.0 | | 分类 | 新闻资讯 | | 状态 | ✅ 已上线 | | 维护者 | FinClaw Core Team | | 最后更新 | 2026-03-19 |
财经新闻与公告Skill，提供热点财经新闻、个股新闻、公司公告、券商研报、业绩预告、新闻情感分析等功能。支持关键词监控，自动识别利好/利空信息。
| 用户输入 | 识别意图 | 调用函数 | |:---|:---|:---| | 有什么财经新闻？ | news_hot | news_hot.py | | 茅台有什么公告？ | announcement | announcement.py 600519 | | 宁德时代最新研报 | research_report | research_report.py 300750 | | 最近有哪些回购公告？ | news_monitor | news_monitor.py 回购 | | 这个新闻是利好吗？ | sentiment_analysis | news_sentiment.py |
| 数据类型 | 主要来源 | 备用来源 | 认证要求 | |:---|:---|:---|:---:| | 热点新闻 | AkShare-东方财富 | 新浪财经 | 无需 | | 公司公告 | AkShare-东方财富 | - | 无需 | | 券商研报 | AkShare-东方财富 | - | 无需 | | 业绩预告 | AkShare-东方财富 | - | 无需 |
bash python scripts/news_hot.py
bash python scripts/news_stock.py 600519 python scripts/news_stock.py 300750
bash python scripts/announcement.py 600519 python scripts/announcement.py 300750 --type 业绩
bash python scripts/research_report.py 600519 python scripts/research_report.py 300750
bash python scripts/news_sentiment.py 600519 python scripts/news_sentiment.py "茅台发布超预期业绩预告"
bash python scripts/news_monitor.py 回购 python scripts/news_monitor.py 增持 python scripts/news_monitor.py 中标
bash python scripts/earnings_preview.py python scripts/earnings_preview.py 600519
| 脚本名 | 功能 | 入口点 | |:---|:---|:---:| | news_hot.py | 热点财经新闻 | ✅ | | news_stock.py | 个股新闻 | ✅ | | announcement.py | 公司公告查询 | ✅ | | research_report.py | 券商研报 | ✅ | | news_sentiment.py | 新闻情感分析 | ✅ | | news_monitor.py | 关键词监控 | ✅ | | earnings_preview.py | 业绩预告 | ✅ |
| 类型 | 说明 | 重要性 | 常见关键词 | |:---|:---|:---:|:---| | 业绩预告 | 季度/年度业绩预测 | ⭐⭐⭐ | 预增、预减、扭亏、预亏 | | 业绩快报 | 业绩快报披露 | ⭐⭐⭐ | 营收、净利润、同比增长 | | 增减持 | 股东增减持计划 | ⭐⭐⭐ | 增持、减持、减持计划 | | 回购 | 股票回购公告 | ⭐⭐⭐ | 回购、注销、股权激励 | | 重大合同 | 重大订单/合同 | ⭐⭐⭐ | 中标、合同、订单 | | 并购重组 | 并购/重组/资产注入 | ⭐⭐⭐ | 重组、收购、资产注入 | | 定向增发 | 定增方案/进展 | ⭐⭐ | 定增、非公开发行 | | 股权激励 | 员工持股/期权激励 | ⭐⭐ | 股权激励、员工持股 | | 分红送转 | 分红、转增股本 | ⭐⭐ | 分红、送转、派息 |
| 类别 | 关键词 | |:---|:---| | 业绩类 | 预增、超预期、扭亏、大增、创新高 | | 资金类 | 增持、回购、员工持股、大资金入场 | | 业务类 | 中标、大单、合同、突破、订单饱满 | | 其他 | 利好、获批、认证、合作、扩张 |
| 类别 | 关键词 | |:---|:---| | 业绩类 | 预亏、预减、下滑、不及预期、亏损 | | 资金类 | 减持、解禁、套现、质押、爆仓 | | 风险类 | 处罚、调查、诉讼、违规、退市 | | 其他 | 利空、停产、裁员、收缩、违约 |
| 分数 | 情感倾向 | 解读 | |:---:|:---:|:---| | +0.8 ~ +1.0 | 强烈积极 | 重大利好，可能涨停 | | +0.3 ~ +0.7 | 积极 | 利好消息，可能上涨 | | -0.2 ~ +0.2 | 中性 | 无明显影响 | | -0.7 ~ -0.3 | 消极 | 利空消息，可能下跌 | | -1.0 ~ -0.8 | 强烈消极 | 重大利空，可能跌停 |
| 监控场景 | 推荐关键词 | 适用对象 | |:---|:---|:---| | 价值投资 | 回购、增持、分红 | 长期投资者 | | 成长投资 | 预增、订单、中标 | 成长股投资者 | | 事件驱动 | 重组、收购、定增 | 短线交易者 | | 风险预警 | 减持、解禁、处罚 | 风险管理者 |
本Skill所有输出数据将按以下格式标注来源：
markdown--- 📊 **数据来源**: AkShare-东方财富 ⏱️ **数据时间**: 2026-03-19 10:30:15 📌 **公告日期**: 2026-03-18 🔗 **原始链接**: 东方财富官网 🔧 **分析工具**: FinClaw v1.0 ⚠️ **免责声明**: 新闻分析仅供参考，不构成投资建议
akshare>=1.10.0
pandas>=1.3.0
jieba>=0.42.0
pyyaml>=5.4.0| 指标 | 目标值 | 当前值 | |:---|:---:|:---:| | 新闻更新延迟 | < 30分钟 | ~15分钟 | | 公告更新延迟 | < 1小时 | ~30分钟 | | 数据可用性 | > 95% | 98% |
| 版本 | 日期 | 变更内容 | |:---|:---:|:---| | 1.1.0 | 2026-03-19 | 符合FinClaw数据规范v1.0，新增情感评分规则 | | 1.0.0 | 2026-03-13 | 初始版本 |
finclaw/config/data_source_config.yaml本Skill遵循 FinClaw 数据规范 v1.0 | 数据来源强制标注 | 禁止训练数据编造
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-03 | fail→pass | 9,130 | 16,696 | +83% | 1 | 1 | 0% | 1,690 | 4,326 | +156% | 0 | 0 | — |
case-01 | fail→pass | 12,634 | 25,368 | +101% | 1 | 1 | 0% | 2,354 | 6,017 | +156% | 0 | 0 | — |
case-02 | fail→pass | 11,590 | 11,796 | +2% | 1 | 1 | 0% | 1,567 | 4,424 | +182% | 0 | 0 | — |
case-04 | fail→fail | 19,928 | 38,587 | +94% | 1 | 1 | 0% | 3,759 | 8,896 | +137% | 0 | 0 | — |
case-05 | fail→fail | 7,937 | 9,269 | +17% | 1 | 1 | 0% | 1,259 | 3,988 | +217% | 0 | 0 | — |
case-06 | fail→fail | 16,368 | 14,789 | -10% | 1 | 1 | 0% | 2,653 | 4,923 | +86% | 0 | 0 | — |
case-07 | fail→pass | 11,929 | 20,433 | +71% | 1 | 1 | 0% | 1,735 | 4,495 | +159% | 0 | 0 | — |
case-08 | fail→fail | 18,923 | 8,049 | -57% | 1 | 1 | 0% | 2,449 | 2,815 | +15% | 0 | 0 | — |
case-09 | fail→pass | 14,828 | 4,260 | -71% | 1 | 1 | 0% | 2,571 | 3,301 | +28% | 0 | 0 | — |
case-10 | pass→pass | 20,184 | 12,662 | -37% | 1 | 1 | 0% | 2,889 | 4,922 | +70% | 0 | 0 | — |
case-11 | fail→pass | 9,473 | 6,732 | -29% | 1 | 1 | 0% | 1,264 | 3,562 | +182% | 0 | 0 | — |
case-12 | pass→pass | 15,815 | 10,683 | -32% | 1 | 1 | 0% | 2,485 | 3,926 | +58% | 0 | 0 | — |
case-13 | pass→pass | 12,497 | 8,620 | -31% | 1 | 1 | 0% | 1,880 | 3,842 | +104% | 0 | 0 | — |
case-14 | fail→pass | 14,803 | 3,322 | -78% | 1 | 1 | 0% | 1,974 | 3,086 | +56% | 0 | 0 | — |
case-15 | pass→pass | 15,953 | 11,129 | -30% | 1 | 1 | 0% | 2,013 | 4,370 | +117% | 0 | 0 | — |
case-16 | fail→pass | 17,785 | 5,011 | -72% | 1 | 1 | 0% | 2,581 | 3,451 | +34% | 0 | 0 | — |
case-17 | fail→pass | 15,622 | 3,224 | -79% | 1 | 1 | 0% | 1,961 | 3,093 | +58% | 0 | 0 | — |
case-18 | fail→pass | 10,328 | 2,418 | -77% | 1 | 1 | 0% | 1,424 | 3,009 | +111% | 0 | 0 | — |
case-19 | pass→pass | 15,770 | 4,276 | -73% | 1 | 1 | 0% | 2,574 | 3,419 | +33% | 0 | 0 | — |
case-20 | fail→pass | 13,638 | 5,149 | -62% | 1 | 1 | 0% | 2,214 | 3,249 | +47% | 0 | 0 | — |
case-21 | fail→pass | 11,767 | 2,925 | -75% | 1 | 1 | 0% | 1,804 | 2,936 | +63% | 0 | 0 | — |
case-22 | fail→pass | 21,282 | 3,552 | -83% | 1 | 1 | 0% | 2,854 | 3,037 | +6% | 0 | 0 | — |
case-23 | fail→pass | 11,197 | 7,042 | -37% | 1 | 1 | 0% | 1,917 | 3,464 | +81% | 0 | 0 | — |
case-24 | pass→pass | 15,829 | 5,017 | -68% | 1 | 1 | 0% | 2,099 | 3,163 | +51% | 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. 24 cases were attempted, and 23 counted toward the lift figure. The other 1 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 +58 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.