Scraper API · JSON Extract exactly what matters as JSON.
Describe the data you need or define a schema, then turn webpages into structured fields ready for your applications.
- A natural language prompt or a JSON schema
- Powered by the general AI agent
- No CSS selectors or XPath parsers to maintain
From URL to structured JSON
Structured JSON based on a prompt or supplied schema, with actual fields depending on the target and extraction configuration.
curl --location --request POST 'https://api.mrscraper.com?token={MRSCRAPER_API_TOKEN}&geoCode=us&html=true&proxyCountry=us' \
-H 'x-api-token: {MRSCRAPER_API_TOKEN}' \
-H 'Content-Type: application/json' \
-d '{
"url": "https://quotes.toscrape.com/",
"prompt": "Extract all data as complete as possible.",
"agent": "general"
}' {
"html": "<!DOCTYPE html>...</html>",
"data": {
"quotes": [
{
"text": "The world as we have created it is a process of our thinking. It cannot be changed without changing our thinking.",
"author": "Albert Einstein",
"author_url": "/author/Albert-Einstein",
"tags": ["change", "deep-thoughts", "thinking", "world"]
},
{
"text": "It is our choices, Harry, that show what we truly are, far more than our abilities.",
"author": "J.K. Rowling",
"author_url": "/author/J-K-Rowling",
"tags": ["abilities", "choices"]
}
],
"next_page": "/page/2/"
},
"event": "done"
} Why teams use it
Less custom parsing
Describe the fields you want instead of writing fragile CSS selectors or custom XPath parsers.
Standardized output
Map different websites into one uniform JSON schema, regardless of the differences in their markup.
Simpler integration with databases and analytics tools
Feed structured payloads into databases, webhooks or downstream ETL pipelines without manual cleanup.
How it works
- 1
Select URL
Send the page URL in the request body, with agent set to general.
- 2
Specify fields or a JSON schema
Describe the data shape in the prompt, or build a JSON schema that is appended to it.
- 3
Run extraction
The general agent analyzes the page against your prompt and returns the fields in data.
- 4
Validate output and integrate
Check the fields in the response, then send them on to your database, webhook or pipeline.
Example use cases
Product title, price and availability
Pull product attributes such as title, price, SKU, availability, variants and image URLs into one schema.
Job listing fields
Turn a single job posting into the fields you name in the prompt, such as title, company and location.
Article metadata
Extract publication dates, author bios, headlines, body copy and category tags from news sites and blogs.
JSON FAQ
Answered from the product documentation.
Should I write a prompt or a JSON schema?
Either works. The prompt describes the data shape you want back, such as title, price or rating. A JSON schema shapes the output further and is appended to the prompt as a "Json Schema:" section when the request is sent.
Can the prompt tell the scraper to click or navigate?
No. The prompt only parses the scraped content. It cannot instruct the scraper itself.
Which AI agent does Scrape JSON use?
The general agent, set with "agent": "general" in the request body. It analyzes the page against your prompt or schema and returns clean, strongly-typed JSON.
Does Scrape JSON follow pagination across several pages?
No. Scrape JSON is optimized for single web pages. For search results, category listings or infinite-scroll catalogs, use the Listing Page preset, which paginates automatically.
What happens if I keep the default prompt?
The default prompt is "Extract all entry data as complete as possible." and the default data type is General. Pick a predefined data type to auto-fill the JSON schema instead.
Explore Our Products
Get page HTML for your own parsing and data workflows.
Turn noisy websites into clean, AI-ready content.
Capture the page as it appears in a browser.
Extract brand assets and style details from a website.
AI-powered endpoints returning structured data from any public site.
Real residential IPs with automatic rotation and global geotargeting.
What people think about
scraper
The mission to make data accessible to everyone is truly inspiring. With MrScraper, data scraping and automation are now easier than ever, giving users of all skill levels the ability to access valuable data. The AI-powered no-code tool simplifies the process, allowing you to extract data without needing technical skills. Plus, the integration with APIs and Zapier makes automation smooth and efficient, from data extraction to delivery.
I'm excited to see how MrScraper will change data access, making it simpler for businesses, researchers, and developers to unlock the full potential of their data. This tool can transform how we use data, saving time and resources while providing deeper insights.
Adnan Sher
Product Hunt user
This tool sounds fantastic! The white glove service being offered to everyone is incredibly generous. It's great to see such customer-focused support.
Harper Perez
Product Hunt user
MrScraper is a tool that helps you collect information from websites quickly and easily. Instead of fighting annoying captchas, MrScraper does the work for you. It can grab lots of data at once, saving you time and effort.
Jayesh Gohel
Product Hunt user
Now that I've set up and tested my first scraper, I'm really impressed. It was much easier than expected, and results worked out of the box, even on sites that are tough to scrape!
Kim Moser
Computer consultant
MrScraper sounds like an incredibly useful tool for anyone looking to gather data at scale without the frustration of captcha blockers. The ability to get and scrape any data you need efficiently and effectively is a game-changer.
Nicola Lanzillot
Product Hunt user
Support
Get help from the MrScraper team with your account, billing, or scraping questions.
Questions? Ask our team via live chat 24/5 or just poke us on our official Twitter or our founder. We're always happy to help.
