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

# Scrape

> Fetch one public page and get back clean markdown plus evidence metadata.

`POST /v1/hosted/scrape`, costs **1 standard credit**.

## Request

Fields on `ScrapeRequest`:

| Field        | Type   | Default        | Notes                                            |
| ------------ | ------ | -------------- | ------------------------------------------------ |
| `url`        | string | ,              | required                                         |
| `formats`    | list   | `["markdown"]` | output formats to return                         |
| `use_js`     | bool   | `false`        | render with a browser instead of plain HTTP      |
| `wait_for`   | string | `""`           | CSS selector or delay to wait for before capture |
| `timeout_ms` | int    | `30000`        |                                                  |
| `stealth`    | bool   | `false`        | enable stealth fetch mode                        |

```bash theme={null}
curl -X POST "$SCOUT_HOSTED_BASE_URL/v1/hosted/scrape" \
  -H "Authorization: Bearer $SCOUT_HOSTED_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"url":"https://example.com","formats":["markdown"],"timeout_ms":30000}'
```

## Response

`ScrapeResponse`, wrapped in hosted usage metadata:

```json theme={null}
{
  "success": true,
  "hosted": {
    "tenant_id": "tn_..",
    "key_id": "key_..",
    "credits_charged": 1,
    "credit_type": "standard"
  },
  "scrape": {
    "success": true,
    "url": "https://example.com",
    "markdown": "# Example Domain\n..",
    "raw_markdown": "..",
    "metadata": {
      "url": "https://example.com",
      "crawled_at": "2026-07-06T18:00:00Z",
      "title": "Example Domain",
      "word_count": 28
    },
    "final_url": "https://example.com",
    "fetched_at": "2026-07-06T18:00:00Z",
    "content_hash": "sha256:..",
    "quality_score": 0.94,
    "duration_ms": 812
  }
}
```

`metadata` and `content_hash` are the evidence block, the source URL, fetch timestamp, and a hash you can diff against future runs.