Structured extraction guide / 2026
Webpage to JSON API with Semantic Entities in 2026
Every site structures data differently. A semantic JSON contract lets an application handle products, posts, profiles, videos, articles, feeds, and ordinary documents without maintaining a parser for every visual layout.
Quick answer
Normalize public pages into one stable JSON schema.
Request format=json to receive schema-v1 output with a semantic entity type, source URL, shared metadata, and type-specific fields. The same contract covers ordinary documents and supported articles, products, posts, profiles, media, and feeds.
curl --get 'https://extractor.sh/api/extract' \
--data-urlencode 'url=https://example.com/' \
--data-urlencode 'format=json'Available data
What you can extract
- Schema version and semantic entity type
- Source URL, title, description, author, dates, and media when available
- Type-specific product, post, profile, audio, video, and feed fields
- Integer minor-unit prices for products and variants
AI workflows
Where normalized data helps
- Typed agent tools
- Knowledge-graph ingestion
- Cross-source normalization
- Structured RAG metadata
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
- Unavailable optional data is not invented.
- Only public page content is represented.
- Applications should validate source freshness and retain canonical URLs.
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.