Como usar
- Copie o comando e verifique o pacote antes de instalar no seu ambiente de agente.
- Consulte no original as ferramentas, credenciais e ambientes necessários.
- Descreva os dados de entrada e a entrega esperada; confira a saída com as fontes.
npx skills add https://github.com/crawlora-org/crawlora-skills --skill news-media-researchAntes de começar
- A instalação não inclui assinatura de modelo, franquia API ou acesso aos seus arquivos.
- Estrelas e instalações indicam atividade, não um teste independente de qualidade ou segurança.
Renderizado do repositório de origem, preservando títulos, exemplos, código, tabelas, links e imagens.
News media research
Search and read public publisher coverage as normalized JSON. This skill is for outlet-native retrieval; use news-briefing-research when the task needs a multi-source synthesis, coverage timeline, or GDELT context.
When to use this skill
- Get the latest headlines for a section, topic, or publisher.
- Read a public article's body and metadata from its canonical URL.
- Search Financial Times or Google News, or browse a publisher's section/topic
archive after discovering its current slugs.
- Inspect BBC/CNN live coverage or Foreign Policy project/live surfaces.
Setup
- Get a Crawlora API key at crawlora.net.
- Set
CRAWLORA_API_KEYin the environment before running the helper. - The helper reads
CRAWLORA_API_KEYfrom the environment and sends requests
to https://api.crawlora.net/api/v1.
Research workflow
- Start with the publisher's discovery surface when one exists: categories,
sections, or topics return the current accepted slugs for the matching headline route. Do not invent section values.
- Fetch a bounded headline/search page, preserve each result's canonical URL,
publication/update time, author, and section, then pass that URL to the publisher's article endpoint when full text is needed.
- Treat feeds as publisher-specific snapshots, not an exhaustive news corpus.
For a comparison across outlets, keep each publisher's result set and timestamps separate before synthesizing.
- Use live-story, project, or conversation routes only for the publisher
surfaces that expose them; they are snapshots, not subscriptions.
Use `reference/endpoints.md` for exact parameters.
Examples
scripts/crawlora.sh /nyt/sections | jq '.'
scripts/crawlora.sh /nyt/headlines section=world | jq '.'
scripts/crawlora.sh /nyt/article url="https://www.nytimes.com/example" | jq '.'
scripts/crawlora.sh /ft/search q="semiconductor exports" sort=date | jq '.'
scripts/crawlora.sh /reuters/news page=1 | jq '.'
scripts/crawlora.sh /foreignpolicy/live | jq '.'Limits
- Public publisher content only; subscriber-only, metered, challenged, video,
podcast, and live-blog surfaces may be unavailable or partial.
- Preserve the source URL and the publisher's timestamps. Do not present a
bounded feed as complete coverage or infer truth from headline agreement.
- Quote sparingly and link the original article. Use
news-briefing-research
for cross-source comparison and uncertainty-aware synthesis.
