Grok Scraper

Collect xAI Grok answers through the Scrapeless LLM Chat Scraper API, including the full response text, follow-up suggestions, web search results, X (Twitter) search results, and source footnotes, without reverse-engineering the Grok UI, maintaining browsers, or building your own anti-blocking stack.

Use this repo when you need a repeatable way to monitor Grok answers for GEO and AI search visibility, compare prompts across regions, audit cited sources and X search results, or pipe AI responses into analytics and automation workflows.

• Full documentation: https://docs.scrapeless.com/en/llm-chat-scraper/quickstart/introduction/
• Get your x-api-token: https://app.scrapeless.com/passport/login?redirect=/quick-start
• API endpoint: POST https://api.scrapeless.com/api/v2/scraper/execute

How it works

Send a single POST request to the Scrapeless endpoint with your API token in the x-api-token header. The body specifies the actor (scraper.grok) and an input object with your prompt and options. The API runs the query and returns the structured result in task_result.

POST https://api.scrapeless.com/api/v2/scraper/execute
Content-Type: application/json
x-api-token: <YOUR_API_TOKEN>

Quick start (curl)

curl 'https://api.scrapeless.com/api/v2/scraper/execute' \
  --header 'Content-Type: application/json' \
  --header 'x-api-token: YOUR_API_TOKEN' \
  --data '{
    "actor": "scraper.grok",
    "input": {
      "prompt": "Most reliable proxy service for data extraction",
      "country": "US",
      "mode": "MODEL_MODE_AUTO"
    }
  }'

To receive the result asynchronously, add a webhook object:

"webhook": { "url": "https://www.your-webhook.com" }

Request parameters

The request body has three top-level fields: actor (always scraper.grok), input (below), and an optional webhook.

Parameter (input.*)	Type	Required	Description
prompt	string	Yes	Prompt to send to Grok.
country	string	Yes	Country / region code (e.g. US, JP).
mode	string	Yes	Model mode: MODEL_MODE_FAST, MODEL_MODE_EXPERT, or MODEL_MODE_AUTO.

Response

A successful call returns a status envelope; the scraped data lives in task_result:

{
  "status": "success",
  "task_id": "e705743d-da2e-4163-9ccd-eef62529ff72",
  "task_result": {
    "conversation": {
      "conversation_id": "...",
      "title": "...",
      "create_time": "...",
      "modify_time": "...",
      "temporary": false
    },
    "user_query": "Most reliable proxy service for data extraction",
    "user_model": "grok-4",
    "full_response": "...answer text...",
    "follow_up_suggestions": [],
    "web_search_results": [
      { "title": "...", "url": "https://...", "preview": "..." }
    ],
    "x_search_results": [
      { "user_name": "...", "name": "...", "text": "...", "url": "https://..." }
    ],
    "tool_usages": [],
    "footnotes": {}
  }
}

Top-level fields

Field	Type	Description
status	string	Request status, e.g. success.
task_id	string	Unique identifier for the task.
task_result	object	Scraped result (fields below).

task_result fields

Field	Type	Description
conversation	object	Conversation metadata (conversation_id, title, create_time, modify_time, temporary).
user_query	string	Original prompt.
user_model	string	Model used, e.g. grok-4.
full_response	string	The answer text from Grok.
follow_up_suggestions	[]string	Suggested follow-up questions.
web_search_results	array	Web search results (title, url, preview).
x_search_results	array	X (Twitter) search results (user_name, name, text, url, post_id, view_count, create_time, profile_image_url).
tool_usages	array	Tools invoked while answering (tool_name, tool_args, card_id).
footnotes	object	Footnotes keyed by ID; each is an object with id, card_type, url.

For the complete field list (nested conversation, footnote, and X search attributes), see the official documentation.

Code examples

Ready-to-run examples live in examples/:

Language	File	Run
Python	example.py	pip install requests && python example.py
Node.js	example.js	node example.js (Node 18+)
Go	example.go	go run example.go
Java	Example.java	java Example.java (Java 11+)
PHP	example.php	php example.php

All examples read the token from the SCRAPELESS_API_TOKEN environment variable:

export SCRAPELESS_API_TOKEN="your_api_token"

Practical use cases

AI answer monitoring

Track how Grok responds to your brand, product category, documentation topics, or competitor prompts. Store the full response text, web search results, X search results, and footnotes so your team can measure AI visibility over time.

GEO and SEO research

Run the same prompt across countries and model modes to compare which sources Grok cites, how recommendations change by region, and where your content appears in AI-generated answers.

Competitor intelligence

Collect structured Grok answers for competitor names, feature comparisons, pricing questions, and "best tool for..." prompts. Use the output, including X (Twitter) search results, to identify messaging gaps and content opportunities.

Dataset and workflow automation

Pipe Grok answers into internal dashboards, knowledge-base QA systems, spreadsheets, data warehouses, or alerting workflows through the synchronous API response or webhook callback.

Why use Scrapeless for Grok scraping?

Benefit	What it means for your team
One unified API	Query Grok through the same Scrapeless LLM Chat Scraper workflow used for other AI answer engines.
Structured output	Receive the full response text, follow-up suggestions, web and X search results, footnotes, and prompt metadata in a developer-friendly response.
Less maintenance	Avoid building browser automation, UI selectors, proxy rotation, retries, and anti-blocking logic yourself.
Region-aware analysis	Use country inputs to compare localized AI answers and source citations.
Production integration	Use API tokens, webhooks, and language examples to connect Grok data to real applications quickly.

FAQ

What is Grok Scraper?

Grok Scraper is a Scrapeless LLM Chat Scraper actor that sends prompts to xAI Grok and returns structured answer data, including the full response text, follow-up suggestions, web search results, X (Twitter) search results, and source footnotes.

Do I need to run a browser or proxy pool?

No. This repo shows how to call the Scrapeless API. Scrapeless handles the scraping workflow behind the API, so your application only needs to send requests and process the returned data.

Which Grok modes are supported?

The current request schema supports MODEL_MODE_FAST, MODEL_MODE_EXPERT, and MODEL_MODE_AUTO. Check the official documentation for the latest supported options before deploying a production workflow.

Can I get results asynchronously?

Yes. Add a webhook object with your callback URL to receive results asynchronously when the task completes.

Is this suitable for AI search visibility monitoring?

Yes. The response includes the full answer text, web search results, X (Twitter) search results, and footnotes, which makes it useful for GEO analysis, brand monitoring, source tracking, and competitive research.

What should I consider before using AI scraping in production?

Make sure your use case complies with applicable laws, platform terms, privacy requirements, and your organization's data policies. Avoid collecting sensitive, private, or unauthorized information.

Learn more

• Scrapeless LLM Chat Scraper documentation
• Supported LLM Chat Scraper actors
• Scrapeless dashboard
• Scrapeless website

Contact us

Need help building a Grok monitoring workflow or scaling AI answer collection?

• Join our Discord.
• Contact us on Telegram.
• For repo-specific issues or improvements, open an issue or pull request in this repository.