Grok Scraper

The Grok scraper by cloro returns Grok answers as structured JSON, with richer source metadata than the other AI surfaces: preview text, site name, author, favicon and image alongside the URL.

How do you scrape Grok?

1. Get an API key at cloro.dev.
2. POST a prompt to https://api.cloro.dev/v1/monitor/grok.
3. Read the parsed fields from the JSON response.

Grok's distinguishing feature for monitoring is recency. It leans on X for current events, so its citations move faster than the other engines and skew toward posts and news rather than evergreen pages. If you are tracking a launch or an incident, this is the surface where it shows up first.

Request sample (Python)

import requests

payload = {
    'prompt': 'what are people saying about the Google num=100 change',
    'country': 'US',
    'include': {'markdown': True},
}

response = requests.post(
    'https://api.cloro.dev/v1/monitor/grok',
    headers={'Authorization': 'Bearer YOUR_API_KEY'},
    json=payload,
)

print(response.json())

Request sample (cURL)

curl -X POST https://api.cloro.dev/v1/monitor/grok \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "latest news on AI search regulation", "country": "US"}'

Node.js and async/webhook examples are in the endpoint documentation.

Request parameters

Parameter	Description	Default
prompt*	The query or question (1-10,000 characters)	–
country	Country code for localized results (US, GB, DE)	US
state	US state code for finer localization	–
include.markdown	Return the answer as Markdown	false
include.html	Return a URL to the full HTML (expires after 24h)	false
include.rawResponse	Return the unparsed upstream payload	false

* Required

What data does the Grok scraper return?

{
  "success": true,
  "result": {
    "text": "Recent discussion centres on the removal of the num=100 parameter...",
    "sources": [
      {
        "position": 1,
        "url": "https://example.com/num100-analysis",
        "label": "Example Analysis",
        "description": "What the num=100 removal changed for rank tracking",
        "preview": "Google removed the num=100 parameter on September 11, 2025...",
        "siteName": "Example",
        "creator": "@example",
        "favicon": "https://example.com/favicon.ico",
        "image": "https://example.com/cover.png"
      }
    ],
    "markdown": "Recent discussion centres on..."
  }
}

Alongside text and markdown:

1. sources — with more metadata per entry than any other cloro surface: preview, searchEngineText, siteName, metadataTitle, creator, image and favicon on top of the usual position, URL, label and description.
2. citationPills — inline citation chips where present.
3. rawResponse — the unparsed upstream payload.

The creator field is why this surface is worth tracking separately: it names the account behind a cited post, which no other engine exposes.

Full field-level schemas are in the endpoint reference.

Use cases

• Real-time news and incident monitoring, where Grok cites faster than the other engines.
• Social listening with attribution — creator names the account behind a cited post.
• Launch tracking — what the model says about a product in the days after announcement.
• Cross-engine comparison — Grok's source mix is the least like the others, so it is a useful outlier check.

FAQ

What makes Grok's sources different?

More metadata per source. Preview text, site name, author, favicon and image come back alongside the URL, so you can render a citation without a second fetch.

Does Grok cite X posts?

Frequently, yes. That is the main reason its citation set diverges from Gemini or ChatGPT on the same prompt.

Is scraping Grok allowed?

cloro reads publicly visible responses. Check your own jurisdiction and terms.

What is the recommended timeout?

60 seconds.

Learn more

• Endpoint reference: cloro.dev/docs
• Product page: cloro.dev/grok

Other cloro scrapers

AI Mode · AI Overview · ChatGPT · Copilot · Gemini · Google Search · Google News · Perplexity

Contact us

Questions or support: r/cloroapi.