> ## Documentation Index
> Fetch the complete documentation index at: https://docs.saturn-pay.com/llms.txt
> Use this file to discover all available pages before exploring further.

# scrape

> URL to structured HTML — raw extraction

The `scrape` capability extracts structured HTML content from URLs, preserving the page structure for data extraction.

## Providers

| Provider   | Features                                      |
| ---------- | --------------------------------------------- |
| Firecrawl  | JavaScript rendering, handles dynamic content |
| ScraperAPI | Rotating proxies, handles blocks              |

## Basic Usage

```typescript theme={null}
const result = await saturn.scrape({
  url: 'https://example.com/products',
});

console.log(result.data.html);
// → Raw HTML content
```

## Parameters

<ParamField body="url" type="string" required>
  The URL to scrape.
</ParamField>

<ParamField body="provider" type="string">
  Specific provider to use (`firecrawl` or `scraperapi`).
</ParamField>

<ParamField body="waitFor" type="number">
  Milliseconds to wait for JavaScript rendering.
</ParamField>

## Response

```typescript theme={null}
interface ScrapeResponse {
  data: {
    html: string;       // Raw HTML content
    url: string;        // Final URL (after redirects)
    statusCode: number; // HTTP status code
  };
  metadata: {
    chargedUsdCents: number;
    provider: string;
    latencyMs: number;
    auditId: string;
  };
}
```

## Examples

### Extract Product Data

```typescript theme={null}
const page = await saturn.scrape({
  url: 'https://store.example.com/product/123',
});

// Use LLM to extract structured data
const extraction = await saturn.reason({
  prompt: `Extract product information from this HTML as JSON:
  - name
  - price
  - description
  - availability

  HTML:
  ${page.data.html}`,
});

const product = JSON.parse(extraction.data.content);
```

### Handle Dynamic Content

```typescript theme={null}
// Wait for JavaScript to render
const page = await saturn.scrape({
  url: 'https://spa.example.com',
  waitFor: 3000, // Wait 3 seconds
});
```

## When to Use `scrape` vs `read`

| Capability | Use case                            | Returns    |
| ---------- | ----------------------------------- | ---------- |
| `read`     | Articles, blog posts, documentation | Clean text |
| `scrape`   | Data extraction, structured content | Raw HTML   |

Use `scrape` when you need to:

* Extract specific elements (prices, tables, lists)
* Parse structured data
* Work with page layout

## Pricing

| Provider   | Cost per page |
| ---------- | ------------- |
| Firecrawl  | \~\$0.002     |
| ScraperAPI | \~\$0.005     |

## Combining with LLM Extraction

```typescript theme={null}
async function extractJobListings(url: string) {
  const page = await saturn.scrape({ url });

  const jobs = await saturn.reason({
    prompt: `Extract job listings from this HTML as JSON array:
    Each job should have: title, company, location, salary (if listed)

    HTML:
    ${page.data.html}`,
  });

  return JSON.parse(jobs.data.content);
}
```