> ## Documentation Index > Fetch the complete documentation index at: https://docs.hydrafetch.com/llms.txt > Use this file to discover all available pages before exploring further. # Extract structured data from URLs > Pull schema-shaped JSON out of one or many pages in a single call. An LLM maps each page onto your schema and/or prompt. Point at a single page, a list, or a crawl scope with a trailing `/*` wildcard; optionally let web search find extra source pages, return per-field confidence with the source passage behind each value, and merge everything into one deduplicated collection of entities. Charged per page that returns data. ## OpenAPI ````yaml https://api.hydrafetch.com/openapi.json post /v1/web/extract openapi: 3.0.0 info: title: Hydrafetch API description: >- Hydrafetch turns any URL into clean, LLM-ready data through one API. Give us a link and get back Markdown, the page's own structured data, extracted JSON, links, a summary, or a screenshot. Point us at a whole site and get every page. Ask a question and get answers with per-field confidence and the passage each value came from. You describe the outcome you want — the pipeline decides how to get it. ## Authentication Every request is authenticated with your API key in the `X-API-Key` header. Keys are scoped to a workspace and carry its credit balance. ## Credits Calls are billed in credits and charged only on success. A standard scrape is one credit; richer formats and the extraction tier cost more. Each response reports what it consumed. ## Conventions All timestamps are UTC ISO 8601. Long-running jobs (crawl, batch) return a job id you poll, or a webhook you register. Errors return a structured body with a stable `code` and an HTTP status. version: '1.0' contact: name: Hydrafetch url: https://hydrafetch.com email: team@hydrafetch.com servers: - url: https://api.hydrafetch.com description: Production security: - apiKey: [] tags: - name: Web Scraping description: >- Turn URLs into clean, LLM-ready content: scrape a single page, crawl or batch-scrape a whole site, map its URLs, search the web, and capture screenshots or images. - name: Web Extraction description: >- Pull schema-shaped JSON out of one or many pages, with optional per-field confidence and the source passage behind each value. paths: /v1/web/extract: post: tags: - Web Extraction summary: Extract structured data from URLs description: >- Pull schema-shaped JSON out of one or many pages in a single call. An LLM maps each page onto your schema and/or prompt. Point at a single page, a list, or a crawl scope with a trailing `/*` wildcard; optionally let web search find extra source pages, return per-field confidence with the source passage behind each value, and merge everything into one deduplicated collection of entities. Charged per page that returns data. operationId: run parameters: [] requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/ExtractRequestDto' responses: '200': description: The extracted data. content: application/json: schema: $ref: '#/components/schemas/ExtractResponseDto' components: schemas: ExtractRequestDto: type: object properties: urls: maxItems: 10 example: - https://example.com/products/widget - https://example.com/products/* description: >- The pages to extract from. Each must be an http(s) URL. A trailing `/*` marks a crawl scope: every page discovered under that path is extracted and merged into the result. type: array items: type: string schema: type: object additionalProperties: true description: >- JSON Schema describing the shape you want back. Optional if `prompt` is given; when both are present the schema fixes the field names and types while the prompt guides what to pull. prompt: type: string maxLength: 2000 example: Pull the product name, price in USD, and whether it is in stock. description: >- Natural-language instruction for what to extract. Use with or instead of a schema. preferStructure: type: boolean description: >- Preserve document structure (headings, lists, tables) over prose density when reading the page — good for listing and catalog pages. Default off. enableWebSearch: type: boolean description: >- Pull in extra source pages by web-searching your prompt, to fill fields your URLs do not cover. Requires a `prompt`. showSources: type: boolean description: >- Return the concrete list of URLs that were actually extracted, after any wildcard and web-search expansion. Default off. showConfidence: type: boolean description: >- For each field, return a confidence score and the exact source passage the value was drawn from. Default off. mergeEntities: type: boolean description: >- Merge the per-page results into one deduplicated collection — one row per entity, with its contributing source URLs — instead of a separate result per page. Default off. maxAge: type: number minimum: 0 maximum: 604800000 description: >- Reuse a recent capture of each page if it is younger than this many milliseconds. Omit or 0 to always fetch fresh. Capped at 7 days. required: - urls ExtractResponseDto: type: object properties: data: $ref: '#/components/schemas/ExtractResultDto' required: - data ExtractResultDto: type: object properties: results: description: One result per extracted page. type: array items: $ref: '#/components/schemas/ExtractItemResultDto' sources: description: >- The concrete URLs actually extracted, after wildcard and web-search expansion. Present only when `showSources` is set. type: array items: type: string collection: description: >- The deduplicated collection, one row per entity. Present only when `mergeEntities` is set. type: array items: $ref: '#/components/schemas/MergedEntityDto' required: - results ExtractItemResultDto: type: object properties: url: type: string example: https://example.com/products/widget description: The page this result came from. data: type: object additionalProperties: true nullable: true description: >- The extracted data, shaped by your schema and/or prompt. Null when nothing matched. fields: type: object additionalProperties: $ref: '#/components/schemas/FieldProvenanceDto' description: >- Per-field confidence and source passage, keyed by field name. Present only when `showConfidence` is set. error: type: object nullable: true example: null description: Set when this page could not be extracted; null on success. required: - url - data - error MergedEntityDto: type: object properties: data: type: object additionalProperties: true description: >- The unioned data for one entity, filled across every contributing page. sources: description: Every source URL that contributed to this entity. type: array items: type: string required: - data - sources FieldProvenanceDto: type: object properties: confidence: type: number minimum: 0 maximum: 1 example: 0.92 description: How certain the value is correct given the page, from 0 to 1. evidence: type: string example: Priced at $49.00 with free shipping. description: >- The exact short passage the value was drawn from, or an empty string if none. required: - confidence - evidence securitySchemes: apiKey: type: apiKey in: header name: X-API-Key ````