Token导航 LogoToken导航TokenDH.com
研究检索操作浏览器clawhub未标认证来源可访问clear审计通过

global-wire全球电讯

Agent Skill

global-wire 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

3,683

周安装

155

GitHub Stars

1

下载量

1,290
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:global-wire(全球电讯)
来源仓库:https://github.com/yly46967-source/global-wire
安装命令:
openclaw skills install global-wire
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install global-wire

简介

将重大世界事件转化为有线简报与滚动时间表格式。

  • 适用于新闻编辑、危机监测与应急响应支持场景。
  • 通过 clawhub 安装,需确认实时事件流与来源验证机制。
  • 输出带来源标注的稳定格式警报条目。global-wire 属于研究检索类 Skill,可作为该场景下的辅助能力补充。
  • 建议设置过滤关键词以匹配特定关注领域。

SKILL.md

name
global-wire
description
Turn answers about major world events into wire-style briefings, rolling event timelines, verification notes, and alert bullets with stable formatting, source priority, and confidence labels. Use when the user asks what happened today, wants a professional global news briefing, needs a conflict or policy timeline, or asks whether a fast-moving report is credible.
metadata
{"openclaw":{"emoji":"📰"}}

GlobalWire

Use this skill when the user wants major world news to read like a professional wire desk instead of a loose AI summary.

Track major world events like a wire desk: professional briefings, credibility labels, and rolling timelines.

GlobalWire is an editorial skill, not a crawler framework.

Default assumption:

  • the current model or active workflow may already have found the core answer
  • GlobalWire's main job is to normalize, tighten, and publish that answer
  • only add extra search when the answer is clearly incomplete or weakly sourced

When the user asks in Chinese, default to fully Chinese section labels and Chinese confidence labels. Do not leave English field names in a Chinese-only response unless the user explicitly asks for bilingual formatting.

Quick start

Example natural-language prompts:

  • 今天发生了什么重大新闻?按 GlobalWire 格式输出。
  • Use $global-wire to summarize today's biggest global events in Chinese.
  • Use $global-wire to build a full timeline of the 2026 US-Iran conflict.
  • Use $global-wire to verify how credible this report is.

Example explicit invocations:

  • $global-wire
  • /global_wire

On surfaces that sanitize slash-command names, the display name remains GlobalWire while the command may appear as /global_wire.

Product promise

GlobalWire should feel like a newsroom editor handling a rough draft:

  • cleaner structure
  • lower noise
  • clearer fact versus analysis separation
  • stronger timeline discipline
  • explicit confidence labels
  • stronger focus on globally consequential developments
  • concise, reusable markdown output

Core modes

Pick one primary mode:

  1. brief

- Use for daily or on-demand major-news summaries. - Return 3-5 independent cards by default, but prefer fewer stronger cards over padding. - Expand to at most 8 on unusually heavy days. - If there is no clear global-level top event, say so and add Still worth watching. - Require clear global spillover for inclusion in the main card list.

  1. timeline

- Use for a continuing event. - Default to the event's full arc from its start to today. - Write event development, not media history.

  1. verify

- Use when the user asks whether a claim or summary is reliable. - Grade reliability; do not pretend to be an absolute fact-checking authority.

  1. alert

- Use for fast, developing, high-impact events. - Keep the first pass short and explicit about uncertainty.

Result-first workflow

  1. Start from the best answer already available in context.
  2. If the answer is obviously thin, add light evidence patching.
  3. Reorganize into the correct GlobalWire mode.
  4. Apply timeline, formatting, source-priority, and confidence rules.
  5. Save to markdown when requested or when the workflow is archival.

Search stance

Do not default to heavy browser automation or multi-platform orchestration.

Preferred order:

  1. current model's native search result or answer
  2. user-provided answer or draft
  3. targeted web search or source checking only when needed
  4. browser automation only as a last-resort patch path

Read {baseDir}/references/source-priority.md before adding or revising sources.

Hard editorial rules

  1. Keep sports, entertainment, local social incidents, and small funding rounds out of the default major-news brief.
  2. Separate Facts from Analysis every time.
  3. Do not write media coverage history as if it were event history.
  4. Do not upgrade weakly sourced timelines into high-confidence narratives.
  5. Domestic sources can supplement, but should not dominate major conflict or geopolitics timelines.
  6. Social posts and prediction markets are weak signals, not factual anchors.
  7. Do not pad with assistant chatter, moral framing, or generic commentary.
  8. Do not pad the daily brief with entertainment, sports, or low-spillover items just to hit a card count target.
  9. Keep confidence labels clean; if nuance is needed, explain it in a note, not by bloating the label.
  10. In Chinese output, localize the briefing header, section labels, confidence labels, and watchlist labels.

References to load

Always use:

  • {baseDir}/references/output-modes.md
  • {baseDir}/references/timeline-rules.md
  • {baseDir}/references/credibility-rules.md
  • {baseDir}/references/source-priority.md
  • {baseDir}/references/formatting-rules.md

Use as needed:

  • {baseDir}/references/exemplars.md
  • {baseDir}/references/file-layout.md

Output contract

Always show:

  • a style line
  • a length line

Default style is always wire unless the user explicitly asks otherwise.

Supported voice modes:

  • wire
  • brief
  • analyst

Translate visible labels to match the user's language while preserving structure and meaning.

Preferred visible labels:

  • English: Current style, Current length
  • Chinese: 当前风格, 当前长度

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

90.48%
按下载量换算1,167

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills