extractor.sh
All guides

News search guide / 2026

News Search API for AI Agents in 2026

News research is most useful when every result includes a publisher, date, summary, and source link. A normalized news feed gives agents those fields without coupling the application to one provider-specific response.

Quick answer

Retrieve current article leads with source attribution.

Send a topic to the news endpoint and receive up to 50 ordered public article results. Use JSON for structured pipelines or Markdown when an AI agent should read the feed directly.

curl --get 'https://extractor.sh/api/news' \
  --data-urlencode 'q=artificial intelligence' \
  --data-urlencode 'limit=10' \
  --data-urlencode 'format=json'

Available data

What you can extract

  • Up to 50 public article results
  • Titles, publishers, dates, summaries, and source links
  • Markdown or provider-neutral schema-v1 JSON
  • Ordered results for a submitted query

AI workflows

Where normalized data helps

  • Current-events research agents
  • News monitoring and classification
  • Source discovery for RAG
  • Publisher and topic analysis

AI-ready output

Markdown for models. JSON for systems.

Raw HTML consumes tokens on navigation, scripts, styling, and interface labels. Clean Markdown keeps the readable hierarchy for LLM prompts and RAG chunks. Normalized JSON is better when your application needs an explicit semantic type, source, author, publication date, media, attributes, and collection items.

Always retain the canonical URL from the response. AI-generated summaries should remain traceable to the public source, especially when the underlying page can change.

Boundaries

Public data only

  • Results contain public article metadata and summaries rather than guaranteed full article text.
  • Coverage and freshness depend on publicly available news results.
  • Sources should be verified before high-stakes decisions.

extractor.sh does not bypass CAPTCHAs, login walls, paywalls, access controls, or regional restrictions. Review the source’s terms and applicable law before collecting or reusing data.