Skip to main content

​
Overview

The options object in a fetch request controls the output format and content of the response. All options are optional — if omitted, the parser returns its default structured output.

​
Options Reference

​
options.markdown
boolean
default:"false"
Return page content as clean, pre-processed Markdown. Strips navigation, ads, and boilerplate HTML. Ideal for LLM consumption.
​
options.json
boolean
default:"false"
Return structured JSON output. For supported platforms (LinkedIn, Amazon, etc.), this returns typed, well-defined fields. For general web pages, returns extracted content in a standard structure.
​
options.include_links
boolean
default:"false"
Extract and include all hyperlinks found on the page in the response.
​
options.include_html
string
Include HTML content in the response. Accepted values:
  • "cleaned" — sanitized HTML with scripts, styles, and non-content elements removed
  • "raw" — original HTML source as-is
​
options.image_urls
boolean
default:"false"
Extract and include all image URLs found on the page.
​
options.json_schema
object
Define a custom schema for structured data extraction. See JSON Schema for details.
​
options.ai
object
AI enrichment options. See AI Enrichment for per-option token costs.

​
Example Request

{
  "url": "https://example.com/article",
  "options": {
    "markdown": true,
    "include_links": true,
    "image_urls": true
  }
}

​
Notes

  • All options are additive — enabling multiple formats returns all of them in a single response
  • Output format selection does not affect token cost (1 token per standard request)
  • AI options are the only options that affect pricing — see AI Enrichment for per-option costs