← 返回 Skills 市场
zhelincheng

Macro News Signal

作者 ZhelinCheng · GitHub ↗ · v1.1.1 · MIT-0
cross-platform ✓ 安全检测通过
210
总下载
1
收藏
0
当前安装
8
版本数
在 OpenClaw 中安装
/install macro-news-signal
功能描述
Macro News Signal is an intelligent market analysis skill that transforms real-time global news and key macro indicators into actionable investment insights.
使用说明 (SKILL.md)

宏观新闻信号

1. 概述

Macro News Signal 是一款智能市场分析工具,旨在通过自动化获取、深度解析、情感计算和多维聚合的工作流程,将碎片化的全球金融新闻与宏观经济指标转化为结构化、可操作的投资决策支持数据。

2.工作流程

新闻请求 → 来源识别 → 自动化获取 → 解析与分析 → 信号生成 → 聚合输出

2.1 第一步:来源识别

根据请求识别合适的新闻来源:

资产类别 主要来源 类型
股票 同花顺、彭博社、CNBC RSS
固定收益 美联储讲话、指数、英央行 RSS/API
大宗商品 EIA、欧佩克、金属公报 网页
外汇 央行、MNI 网页
一般宏观 华尔街日报、金融时报、经济学人、联合早报 RSS
股票指数 Yahoo Finance API

2.2 第二步:数据获取(请严格遵守下面的方案)

所有数据来源均存在 references/news_apis.md 中。

2.2.1 RSS订阅源、指数接口 请求方式

在获取RSS订阅源、指数接口时,需要先判断是否存在curl命令时,如果存在优先使用curl进行数据获取,示例如下:

curl '地址' \
  -H 'accept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7' \
  -H 'accept-language: zh-CN,zh;q=0.9' \
  -H 'cache-control: no-cache' \
  -H 'user-agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/146.0.0.0 Safari/537.36'

当不存在curl命令时,使用默认方式进行数据获取,例如web_fetch。

2.2.2 股票接口 请求方式

当请求股票接口 Yahoo Finance API 时,必须使用 scripts/format.py 对 Yahoo Finance API 原始响应进行格式化,转化为 AI 友好的线性数据结构:

python3 scripts/format.py "\x3Capi_url>"

详细 API 端点和数据格式参见 references/news_apis.md 中的 Yahoo Finance API 相关说明。

2.3 第三步:解析与分析

该阶段利用自然语言处理 (NLP) 提取非结构化文本中的核心变量:

  1. 实体提取 (NER): 自动识别新闻中提到的特定资产(如 $NAS100$)、经济指标(如 $CPI$、$TIPS$)及关键人物或地理区域。
  2. 情感极性标注:
  • 鹰/鸽分析 (Hawkish/Dovish): 针对央行沟通,量化政策偏向。
  • 利好/利空 (Bullish/Bearish): 基于金融词典计算文本情感得分 $S$。
  1. 预测值比对: 若新闻涉及经济数据发布,自动对比“实际值”与“预期值”,计算超预期偏差。

2.4 第四步:信号生成

将解析后的分析转化为量化的投资逻辑:

  • 冲击等级: 划分为 Flash (瞬时波动)、Secondary (次要影响) 或 Trend-Setting (趋势级信号)。
  • 指标相关性: 宏观指标,计算当前新闻对特定资产(如黄金与 10Y TIPS 收益率背离)的映射强度。
  • 逻辑校验: 自动检测是否存在“利好出尽”或“情绪过热”的背离信号。

2.5 第五步:聚合

按以下多维方式聚合分析结果,生成结构化报告:

  • 时间窗口: 每日综述、每周深度回顾。
  • 核心主题: 通货膨胀、GDP、就业市场、央行动态、地缘政治。
  • 地区: 美国、中国、欧盟、新兴市场。
  • 资产类别建议:
    • 买入 (Buy): 强利好信号且情绪合理。
    • 持有 (Hold): 信号中性或处于数据真空期。
    • 卖出 (Sell): 结构性利空或情感极度亢奋。

输出格式: 生成的报告必须严格遵循 references/output_format.md 中定义的模板结构,包括信号级别定义、资产信号定义和冲击强度定义。

3. 定义

3.1 信号级别定义

级别 说明 持续时间
Trend-Setting 趋势级,影响未来数周至数月 长期
Sustained 持续级,影响未来数天至数周 中期
Flash 瞬时冲击,仅当时段有效 短期

3.2 资产信号定义

信号 含义
📈 买入 预计上涨,适合做多
📉 卖出 预计下跌,适合做空
➡️ 持有 预期震荡,适合观望

3.3 冲击强度定义

强度 说明
极高 重大黑天鹅,对市场有决定性影响
重要事件,能引发显著市场波动
常规事件,可能引发短期波动
次要事件,市场影响有限

4. 资源

4.1 references/

文件 内容
news_apis.md 新闻API文档、RSS订阅源、指数接口及 Yahoo Finance API
output_format.md 报告输出格式模板

4.2 scripts/

文件 内容
format.py Yahoo Finance API 数据格式化脚本,将嵌套 JSON 转为扁平化结构
安全使用建议
This skill appears to do what it says and doesn't ask for secrets or installs. Before enabling: 1) remember it will make outgoing HTTP requests to many third‑party news/API endpoints — ensure that is allowed in your environment and complies with terms of use; 2) avoid supplying internal/private URLs as inputs (scripts/format.py will fetch any URL you give it, which could be abused to access internal services); 3) if you need stronger isolation, run the skill in a network-restricted environment or review/modify the script to whitelist only intended endpoints; 4) verify you trust the listed RSSHub/Folo endpoints and Yahoo Finance usage for your compliance needs.
能力评估
Purpose & Capability
Name/description match the actual behavior: the skill collects public news/APIs (RSS, web pages, Yahoo Finance), parses text, produces sentiment/indicator-based signals and structured reports. Required resources (none) align with a lightweight, instruction-driven aggregator.
Instruction Scope
Runtime instructions direct the agent to fetch many third-party endpoints (RSS feeds, public APIs, web pages) which is expected. The included scripts/format.py will fetch and parse any API URL passed on the command line (it accepts an arbitrary URL and issues an HTTP request), so a user/agent-provided URL could cause requests to unintended internal or private endpoints (SSRF-like risk) if misused. The skill does not instruct reading local files or exporting data to unknown endpoints; format.py prints results to stdout only.
Install Mechanism
No install spec and no external packages or downloads are required; the skill is instruction-only with a small Python utility using only the standard library — low install risk.
Credentials
No environment variables, credentials, or config paths are requested. All required external accesses are public news/APIs discussed in the docs, which is proportionate to the stated purpose.
Persistence & Privilege
The skill does not request always:true, does not modify other skills or system settings, and is user-invocable only. Autonomous invocation is allowed by platform default but not combined with other privilege escalations here.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install macro-news-signal
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /macro-news-signal 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.1.1
Version 1.1.1 of macro-news-signal - No file changes detected in this version. - No functional or documentation updates present. - This release maintains all features and documentation as in the previous version.
v1.1.0
- Added scripts/format.py to format Yahoo Finance API responses into AI-friendly linear data. - Removed references/data_schema.md; data schema details are now omitted. - Updated workflow and documentation to require use of format.py for Yahoo Finance API data processing. - Cleaned and clarified acquisition methods for different data sources in SKILL.md. - Revised resource references to reflect the updated file structure and new workflow.
v1.0.5
- No changes detected from the previous version; no file changes were made in this release.
v1.0.4
Version 1.0.4 Changelog - Added explicit definitions for "信号级别", "资产信号", and "冲击强度", clarifying report output classification. - No functional or workflow changes; documentation expanded to include detailed structure and explanation for report signal levels and asset recommendations. - Maintained prior workflow and resource references for clarity and completeness.
v1.0.3
- Added references/output_format.md to define the required report output format. - SKILL.md: Updated news获取流程,curl未可用时可回退到web_fetch。 - SKILL.md: 报告输出须严格遵循 references/output_format.md 的结构和模板。
v1.0.2
- Added requirement to read robots.txt and adhere to website crawl protocols when performing web scraping. - Clarified that if agent-browser is unavailable, use a default fetch method such as web_fetch, and emphasized appropriate request rates for web scraping tasks.
v1.0.1
Version 1.0.1 - Clarified curl request instructions in the data fetch workflow, specifying strict adherence to request conventions. - Updated example curl command to reference endpoints from `references/news_apis.md` directly. - No changes to logic or features; documentation wording only.
v1.0.0
Initial release – Macro News Signal v1.0.0 - Transforms real-time global news and macro indicators into actionable investment insights. - Automates source identification, news retrieval (via RSS, API, or browser), and parsing. - Uses NLP for entity extraction, sentiment analysis (hawkish/dovish, bullish/bearish), and surprise measurement on economic releases. - Generates quantifiable investment signals based on news impact, correlation, and logical validation. - Aggregates results into structured reports by time, theme, geography, and asset class recommendation.
元数据
Slug macro-news-signal
版本 1.1.1
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 8
常见问题

Macro News Signal 是什么?

Macro News Signal is an intelligent market analysis skill that transforms real-time global news and key macro indicators into actionable investment insights. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 210 次。

如何安装 Macro News Signal?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install macro-news-signal」即可一键安装,无需额外配置。

Macro News Signal 是免费的吗?

是的,Macro News Signal 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Macro News Signal 支持哪些平台?

Macro News Signal 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Macro News Signal?

由 ZhelinCheng(@zhelincheng)开发并维护,当前版本 v1.1.1。

💬 留言讨论