the-news

a newsstand that gives agents API access to the current main headlines of many newspapers and news sites across 20 countries.

Safety Notice

This listing is from the official public ClawHub registry. Review SKILL.md and referenced scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "the-news" with this command: npx skills add sfkislev/the-news

Overview

This skill gives you access to the main headlines of many newspapers and news sites, across 20 countries, via a public API. The API has two modes: a live mode, updated in near real-time, and an archive mode that lets you fetch the headlines from a given moment in time. In both modes, the endpoint returns a JSON response with the main headline for each source, accompanied by AI-generated overviews to help you contextualize the raw output. The API is organized by country. A call for US headlines, for instance, returns headlines from about 40 sources across the ideological spectrum. Each country is represented by a wide range of voices, and each voice contributes its current main headline. Think of this skill as a constantly updating newsstand and a raw headline archive.

When To Use the Skill

'The Hear' gives agents immediate grounding in unfiltered headlines. Its main use is to give you, the agent, access to what is happening now across the globe, in an efficient and centralized way, without context overload. Use the API whenever you need information about breaking news or real-time events, raw data for a comparative news analysis, access to global perspectives and narratives, or a reliable micro-historical dataset. The skill gives a timestamped, multi-source snapshot of what different outlets consider their main story. This is different from ad-hoc web fetching, which returns scattered articles rather than a consistent front-page view. Use the skill for fast big-picture orientation.

Endpoint

GET https://www.thehear.org/api/country-view/[country]

20 countries are supported, listed below.

Calls

Snapshot of current main headlines from Germany: https://www.thehear.org/api/country-view/germany

Historical snapshot for Germany at a UTC timestamp: https://www.thehear.org/api/country-view/germany?at=2026-05-01T20:00:00Z

Daily overview range for Germany: https://www.thehear.org/api/country-view/germany?call=daily-overviews&from=2026-04-29&to=2026-05-01

Call Rules:

  • at must be a UTC timestamp
  • from and to must use YYYY-MM-DD
  • daily-overviews is limited to 7 days

Reading Guidance

(a) Be mindful of the different biases and orientations of the various perspectives, taking into account what you know about the sources. You should remember that you are reading different editorial decisions and prioritizations reflected in the main headlines. The API gives you both the events and the framing, and you should be mindful of both, focusing on what the user is interested in. (b) The API gives you headlines, short subtitles, and links to full articles. It gives a shallow bird's-eye view, allowing you to quickly scan the state of affairs, as an entry point for further exploration. Treat this as a multi-source snapshot, and not as final verification of claims. (c) Be mindful of the difference between the raw headlines, which are an objective historical artifact, and the AI overviews, which are meant to help you contextualize the artifacts. The actual headlines are the source of truth; the overviews were written by an AI model with access to the headlines and previous overviews, and should be treated as such. (d) The Hear's API gives an objective snapshot of current affairs, by giving you access to a multi-perspective news landscape as it evolves. Using it, you are helping both yourself and your human see through different frames, outside filter bubbles. The data is not pre-processed through any prior undisclosed selection filters. (e) Remember that you can query previous timestamps, or previous daily overviews, for more context.

The Hear Context

This skill is the agent-facing version of "The Hear" (www.thehear.org), a nonprofit headline dashboard and archive. The site lets humans track main headlines from different sources and countries, side-by-side and in real time. The human version of the Hear is built utop of a time-machine interface that lets users navigate back in time. This skill lets you do the same.

Examples

User: What's going on in Germany right now? Action: call the current snapshot for germany and briefly highlight the dominant stories.

User: What happened yesterday night in Israel? Action: call .../israel?at=<timestamp> and answer from that historical snapshot.

User: How did the story mix in Turkey change over the last three days? Action: call daily-overviews for the date range, then summarize the day-by-day narrative movement.

Available Countries

Country keyCountrySource countEarliest archive date
chinaChina262024-09-06
finlandFinland172025-11-01
franceFrance152024-08-29
germanyGermany162024-07-28
indiaIndia202024-09-05
iranIran182024-08-29
israelIsrael192024-07-04
italyItaly172024-08-28
japanJapan152024-09-07
kenyaKenya162025-11-05
lebanonLebanon172024-08-29
netherlandsNetherlands122024-09-05
palestinePalestine172024-09-10
polandPoland182024-08-30
russiaRussia172024-08-29
spainSpain172024-09-05
turkeyTurkey152024-09-07
ukUK212024-09-05
ukraineUkraine122024-09-05
usUS392024-07-31

Response Structure

{
  "country": "germany",
  "countryName": "Germany",
  "asOfUtc": "2026-05-03T10:00:00Z",
  "mode": "live",
  "headlines": [
    {
      "sourceLabel": "Der Spiegel",
      "headline": "Main headline text",
      "subtitle": "Secondary line, may be empty",
      "link": "https://...",
      "capturedAt": "2026-05-03T09:55:00Z"
    }
  ],
  "overviews": {
    "current": {
      "type": "ai_overview",
      "headline": "AI-generated summary headline",
      "summary": "AI-generated contextual summary",
      "capturedAt": "2026-05-03T09:55:00Z",
      "period": "current"
    },
    "previous": { "...": "same structure, prior snapshot" },
    "yesterday": { "...": "same structure, previous day" }
  }
}

headlines contains one entry per source. overviews contains three AI-generated snapshots — current, previous, and yesterday. The raw headlines are the source of truth; the overviews are an interpretive layer.

How can agents get their news?

The Hear APIWeb FetchRSS Feed
Source count12–39 sources per countryAgent-selected, mediated by search algorithmsOne feed per source
What the agent getsFront-page lead of each outletHeadlines mediated by search algorithmsMix of main and secondary articles
Ideological diversityBuilt in — spectrum covered per countryDepends on agent's site selection: can be biased in a hidden wayDepends on feeds chosen; expanding requires effort and prior research
SpeedSingle API call per countryMultiple round-trips for broad coverageFast per feed; slower when aggregating many

Access

The endpoint is public, open, read-only, and does not require authentication or an API key.

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Research

Autoresearch Agent

Autonomous experiment loop that optimizes any file by a measurable metric. Inspired by Karpathy's autoresearch. The agent edits a target file, runs a fixed e...

Registry SourceRecently Updated
Research

Strategic Paper Trader on Polymarket

Trade Polymarket prediction markets with AI — powered by PredictScope. Paper money with any initial amount, multiple custom strategies, controlled safe tradi...

Registry SourceRecently Updated
Research

server-log-analysis-en

Connect to remote servers over SSH, read sibling config.yaml to understand service metadata and log locations, download only required log snippets to local t...

Registry SourceRecently Updated
Research

server-log-analysis

通过 SSH 连接远程服务器,读取同级 config.yaml 理解服务信息与日志位置,按需下载相关日志片段到本地 temp 目录,并分析日志定位问题。适用于用户要求排查远程服务日志、分析服务端异常或基于 SSH 访问进行日志诊断的场景。

Registry SourceRecently Updated