
Website Content Crawler
scrape.ly/store/querystone/website-content-crawlerAIDeveloper tools
1 monthly user

Send queries to ChatGPT with web search and get the answer, cited sources, recommended products and brand mentions. Built for GEO and AI visibility tracking. No login needed.
Send your questions to ChatGPT with web search turned on and get back the full answer, every source it cites, the products it recommends and how often your brands show up. Built for GEO (generative engine optimization) and AI visibility tracking. No ChatGPT account, no API key, no login.
| Field | Description |
|---|---|
queries | Prompts to send to ChatGPT, one per line. Each line is one result. |
brands | Optional brand or product names to track in each answer. |
countryCode | Country the questions are asked from (default us). |
webSearch | Turn on ChatGPT's web search for every query (default true). |
includeRawMarkdown | Also save the answer exactly as ChatGPT streamed it (default false). |
maxConcurrency | Parallel browser sessions (default 3, up to 8). |
{
"queries": "What are the top 5 CRM tools\nbest project management software for small teams",
"brands": ["HubSpot", "Salesforce", "Pipedrive"],
"countryCode": "us"
}
| Field | Description |
|---|---|
query | The prompt that was sent. |
text | ChatGPT's answer as clean Markdown. |
sources | Sources with id, title, url, domain, section (citations, attached or more), sourceName and publishedAt. |
citationsCount | Number of inline citations. |
citedDomains | Cited domains in order of appearance. |
products | Recommended products with name, price and merchants (when ChatGPT shows products). |
brandMentions | Per tracked brand: mentioned, mentions, rank (order of first mention), cited, citedUrls. |
brandsMentioned | Tracked brands found in the answer, in order of first mention. |
usedWebSearch | Whether the answer cites web sources. |
countryCode, answerSeconds, answerComplete, scrapedAt | Run details. |
{
"query": "What are the top 5 CRM tools",
"text": "The **top 5 CRM (Customer Relationship Management) tools** depend on your business size, budget, and sales process. For most businesses, the strongest options are Salesforce, HubSpot, Microsoft Dynamics 365, Zoho CRM, and Pipedrive.\n\nA CRM helps you manage customer information, track leads and sales opportunities, automate follow-ups, and improve customer relationships.\n\n## Top 5 CRM tools\n\n### 1. Salesforce\n\n(Best f...",
"sources": [
{
"id": "1",
"title": "Best CRM Software for 2026",
"url": "https://technologyadvice.com/crm/",
"domain": "technologyadvice.com",
"section": "citations",
"sourceName": "TechnologyAdvice",
"publishedAt": "October 2, 2026"
},
{
"id": "2",
"title": "CRM Systems Comparison: Which Platform Fits Your Team?",
"url": "https://techyorker.com/crm-systems-comparison-salesforce-hubspot-dynamics-365-zoho-pipedrive-and-freshsales/",
"domain": "techyorker.com",
"section": "citations",
"sourceName": "TechYorker",
"publishedAt": "August 14, 2026"
},
{
"id": "9",
"title": null,
"url": "https://linkgo.dev/tools/salesforce-mcps-2025-10-06",
"domain": "linkgo.dev",
"section": "more"
}
],
"sourcesCount": 23,
"citationsCount": 3,
"citedDomains": [
"technologyadvice.com",
"techyorker.com",
"techradar.com"
],
"products": [],
"usedWebSearch": true,
"countryCode": "us",
"answerComplete": true,
"answerSeconds": 9.9,
"scrapedAt": "2026-10-15T21:20:39+00:00",
"brandMentions": [
{
"brand": "HubSpot",
"mentioned": true,
"mentions": 7,
"firstPosition": 175,
"citedUrls": [
"https://www.hubspot.com/"
],
"rank": 2,
"cited": true
},
{
"brand": "Salesforce",
"mentioned": true,
"mentions": 6,
"firstPosition": 163,
"citedUrls": [
"https://www.salesforce.com/"
],
"rank": 1,
"cited": true
}
],
"brandsMentioned": [
"Salesforce",
"HubSpot",
"Pipedrive"
]
}
You pay $0.15 per answer saved to the dataset ($150 per 1,000 answers). Failed queries are never charged.
On top of that you pay the platform usage of the run: the hosted browser session and a little compute. In our tests that came to about $0.01 to $0.02 per answer.
Example: 100 prompts ≈ $15 in answers + about $1.50 usage ≈ $16.50.
Set a maximum cost per run and the actor stops cleanly before going over it.
Is it legal? The actor reads the answers ChatGPT shows any visitor without an account. You are responsible for how you use the data and for following the terms that apply to you.
Do I need a ChatGPT account, an API key or a proxy? No. Everything runs in a hosted browser session without logging in.
Why are answers different between runs? ChatGPT's answers change from run to run, like on chatgpt.com. For GEO tracking, run the same prompts several times or on a schedule and compare the share of mentions.
What if a query fails? The actor retries in a fresh session. Queries that still fail are listed in the run's OUTPUT record and are not charged.
Can I schedule it or use the API? Yes. Schedule runs in the console or start them through the API and read the dataset as JSON, CSV or Excel.
Which ChatGPT model answers? The model ChatGPT gives logged-out visitors. You cannot choose the model.
Every field below can be set in the run form, passed through the API, or saved as a task. Unset fields fall back to the defaults shown.
queriesstringRequiredQueries
The prompts to send to ChatGPT, one per line. Each line becomes one result. Duplicates are skipped.
Track how often brands appear in ChatGPT's answers (GEO / AI visibility).
brandsarrayOptionalBrands to track
Optional. Brand or product names to look for in each answer. Every result then shows whether each brand is mentioned, how often, in which order and whether its website is cited.
countryCodeenumOptionaldefault "us"Country
Country the questions are asked from. ChatGPT tailors shops, prices and local sources to it.
webSearchbooleanOptionaldefault trueForce web search
Turn ChatGPT's web search on for every query, so answers cite live sources. Turn it off to see what ChatGPT answers from its own knowledge (it may still search on its own).
includeRawMarkdownbooleanOptionaldefault falseInclude raw answer markup
Also save the answer exactly as ChatGPT streamed it, including its layout components and citation markers.
maxConcurrencyintegerOptionaldefault 3Parallel browser sessions
How many queries are asked at the same time. Higher is faster for long lists.
Every run writes its results to a dataset. Preview it in the dashboard, then export it or read it through the API.
| Field | Type | Description |
|---|---|---|
| query | string | The query sent to ChatGPT. |
| text | string | ChatGPT's answer as clean Markdown. |
| sources | array | Sources in the answer: section 'citations' (cited inline), 'attached' (links in the answer) or 'more' (other pages ChatGPT used). |
| sourcesCount | number | |
| citationsCount | number | |
| citedDomains | array | |
| products | array | |
| brandMentions | array | |
| brandsMentioned | array | |
| usedWebSearch | boolean | |
| countryCode | string | |
| answerComplete | boolean | |
| answerSeconds | number | |
| scrapedAt | string | |
| rawMarkdown | string |
You pay for the events the actor reports, such as each result or page it produces. Platform usage is included in the event prices.
| Event | Charged when | Price |
|---|---|---|
| ChatGPT answer | Charged once for each ChatGPT answer saved to the dataset (text, sources and brand mentions included). Failed queries are not charged. | $0.15 |
Run this actor from your own code. Swap in your API token and the examples work as-is.
import { ScrapelyClient } from 'scrapely-client'
const client = new ScrapelyClient({ token: process.env.SCRAPELY_TOKEN })
// Start the actor and wait for the run to finish.
const run = await client.actor('querystone/chatgpt-search-scraper').call({
"queries": "What are the top 5 CRM tools\nbest project management software for small teams",
"brands": [
"HubSpot",
"Salesforce",
"Pipedrive"
]
})
// Read the dataset the run produced.
const { items } = await client.dataset(run.defaultDatasetId).listItems()
console.log(items)Tokens live in your account settings. Keep them server-side: a token carries the full permissions of your account.
Connect the Scrape.ly MCP server once and your agent can run this actor as a tool. Pick your tool:
Fetch scrape.ly/agents.md and run querystone/chatgpt-search-scraperPaste into Claude Code, Codex, Cursor or any AI agent. It signs you in once in the browser.
{
"mcpServers": {
"scrapely": {
"url": "https://api.scrape.ly/mcp?tools=actors,runs,storage,docs&actors=querystone/chatgpt-search-scraper"
}
}
}No key in the file: your client opens the browser to sign you in the first time it connects.
Inputs the developer prepared for common jobs. Open one to see its input, then copy it to your account to run it as-is or adjust it.
ChatGPT web answer with brand mentions and sources.
Answers in Markdown with brand mention checks.
Scrape.ly gives you the infrastructure, the tooling, and the team to extract, process, and operationalize web data at any scale. Setup takes minutes. The advantage lasts forever.