Firecrawl Scrape

Firecrawl Scrape on xpay — clean content from any URL, with optional schema extraction

firecrawl_scrape fetches a single URL, runs full browser rendering for JavaScript-heavy pages, and returns the result in your chosen format: clean Markdown, raw HTML, screenshot, links, or structured JSON conforming to a schema you provide. It's the workhorse Firecrawl endpoint and the right default for single-page content extraction.

Output formats

• markdown (default) — reading-quality Markdown. Best for LLM consumption.
• html / rawHtml — cleaned or raw HTML.
• links — every URL discovered on the page.
• screenshot — full-page or above-the-fold PNG.
• json — schema-driven structured extraction (the killer feature).

Schema-driven JSON extraction

Pass a JSON Schema in the request and Firecrawl runs the page, finds the data points, and returns typed JSON that matches your schema. No regex, no CSS selectors, no DOM debugging.

{
  "url": "https://anthropic.com/pricing",
  "formats": ["json"],
  "jsonOptions": {
    "schema": {
      "type": "object",
      "properties": {
        "tiers": {
          "type": "array",
          "items": {
            "type": "object",
            "properties": {
              "name": {"type": "string"},
              "input_per_mtok": {"type": "number"},
              "output_per_mtok": {"type": "number"}
            }
          }
        }
      }
    }
  }
}

You get back { "tiers": [{ "name": "Sonnet", "input_per_mtok": 3, "output_per_mtok": 15 }, ...] }. This is materially better than scraping the page and asking an LLM to extract JSON — Firecrawl handles JS rendering, retries, and the extraction in one billable call.

Key options

When firecrawl_scrape is right

• You know the URL and want clean content or typed JSON from it.
• The page is JS-rendered and a plain HTTP fetch returns empty HTML.
• You need structured fields without writing scrapers.

When something else is better

• Multiple URLs from the same domain — firecrawl_crawl is more efficient.
• You don't know which page has the data — firecrawl_search or firecrawl_map first.
• High-volume static-HTML scraping — Bright Data batch tools are cheaper.

Pricing

firecrawl_scrape is $0.01–$0.05 per call on xpay depending on options (JSON extraction and screenshots cost a bit more). $5 free credit on signup.