Structured Web ExtractionService

Fetches a web page and returns structured data from it. POST JSON {"url": "https://..."} to /v1/url. Without a schema, fields holds the page standard fields (title, description, canonical, language, h1, OpenGraph og_title, og_description, og_image, og_type, site_name, author, published) and page holds headings (h1-h3), JSON-LD blocks, links_count and a text excerpt. With "schema": ["field", ...] or a JSON Schema, each field is taken from the standard fields or from a "Label: value" line in the page text; with {"fields": {"name": "<regex>"}} each regex is applied to the page HTML. JavaScript is not executed. Pages are cached 10 minutes, so results can change over time.