Search

Web Scrape

Fetch one public web page and return its title, meta data, visible text and links.

live web.scrape $0.0007 per request ($0.70 per 1,000)

Pricing

$0.0007 per request ($0.70 per 1,000), paid from a prepaid USD wallet. The exact amount is on every response as cost_usd; max_cost_usd refuses anything higher; failed calls are refunded.

1,000 requests = $0.70 · 10,000 = $7.00 · 100,000 = $70.00

Run it

curl -s -X POST https://getallapis.com/v1/run/web.scrape \
  -H "Authorization: Bearer $GETALLAPIS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"url":"https://example.com"}'
from getallapis_client import Client
client = Client()  # reads GETALLAPIS_API_KEY
result = client.run('web.scrape', {'url': 'https://example.com'})
print(result['cost_usd'], result['output'])
getallapis run web.scrape '{"url":"https://example.com"}'

Quote first, free: POST /v1/quote/web.scrape with the same body. MCP: run_endpoint with endpoint="web.scrape".

Input

FieldTypeRequiredDefaultNotes
urlstringyesPublic http(s) URL
include_htmlbooleannofalseAlso return the raw HTML (up to 2 MB)
include_linksbooleannotrue

Output

Returned in output of the run envelope (ok, request_id, endpoint, status, cost_usd, items, latency_ms, source, output).

FieldMeaning
final_urlURL after redirects.
statusHTTP status.
titlePage title.
descriptionMeta description.
textVisible text (max 100k characters).
linksAbsolute links on the page (max 500).
htmlRaw HTML when include_html=true.

Source and latency

GetAllAPIs fetcher (first party). Own HTTP fetcher; public pages only, robots.txt honoured, SSRF-guarded. One upstream per endpoint; no automatic failover.

Not enough calls measured yet.

Markdown reference · JSON · Playground