Extract - Nimble Docs

Looking for the easiest way to get web data? Web Search Agents handle everything automatically — no selectors needed. Try Nimble Studio to create an agent for any website in seconds.Extract is for developers who need full control over CSS selectors, browser actions, and page interactions.

Nimble Extract retrieves and parses content from any URL, giving you clean structured data instead of raw HTML. Point it at a webpage, specify what you want, and get back exactly the information you need - with full JavaScript rendering and anti-bot protection.

Quick Start

Example Request

from nimble_python import Nimble

nimble = Nimble(api_key="YOUR-API-KEY")

result = nimble.extract(
    url= "https://www.google.com/search?q=nimble",
    render= True
)

print(result.data.html)

Example Response

{
  "url": "https://www.example.com",
  "task_id": "b1fa7943-cba5-4ec2-a88c-4d2d6799c794",
  "status": "success",
  "data": {
    "html": "<!DOCTYPE html><html>...</html>",
    "headers": {}
  },
  "metadata": {
    "query_time": "2026-02-08T22:00:36.132Z",
    "query_duration": 1877,
    "driver": "vx8"
  },
  "status_code": 200
}

How it works

You provide a URL and options

Give Extract the webpage URL and configure rendering, parsing, or browser actions

Extract fetches and renders the page

Loads the webpage with optional JavaScript rendering if enabled - Handles cookies, headers, and authentication - Bypasses anti-bot protections with stealth mode - Renders dynamic content completely

Extracts your specified data

Returns content in your chosen format (HTML, markdown) - Parses structured data using your CSS selectors (if parsing schema provided) - Captures network requests if configured

Returns structured results

Get clean JSON with your extracted data, ready to use in your application

Parameters

Supported input parameters:

url - Required

url

string

required

The webpage URL to extract content from.Example: https://www.example.com/product

render

boolean

default:"false"

Enable JavaScript rendering for dynamic content.Set to true for sites built with React, Vue, Angular, or other JavaScript frameworks.

driver

string

default:"vx6"

The extraction engine to use.Options:

vx6 - Fast HTTP requests (no JavaScript)
vx8 - Headless browser with JavaScript
vx8-pro - Headful browser with JavaScript
vx10 - Stealth headless browser
vx10-pro - Stealth headful browser

formats

array

Output formats to return.Options:

html - Raw HTML content
markdown - Converted markdown format

Example: ["html", "markdown"]

country

string

default:"ALL"

Extract content as if visiting from a specific country. Use ISO Alpha-2 codes.Examples: US, UK, DE, FR

state

string

For US locations, you can specify a specific state. Only works in US or CA.Use ISO Alpha-2 codes like NY, FL, etc.

Example:

"state": "CA"

city

string

Target a specific city for hyper-local content. Works with most major cities worldwide.

Example:

"city": "New York"

Replace spaces from city names with underscore (e.g,New York becomes new_york).

locale

string

default:"auto"

Set the browser’s language preference. Affects how websites display content to you.Use LCID standard codes like en-US, en-GB, fr-FR, de-DE, etc.

Example:

"locale": "en-US"

parsing

object

CSS selectors for structured data extraction. Define field names with selectors and optional types.Example:

{
    "parser":
    {
        "product_name":
        {
            "type": "terminal",
            "selector":
            {
                "type": "css",
                "css_selector": ".product-title"
            },
            "extractor":
            {
                "type": "text"
            }
        },
        "price":
        {
            "type": "terminal",
            "selector":
            {
                "type": "css",
                "css_selector": ".price-value"
            },
            "extractor":
            {
                "type": "text",
                "post_processor":
                {
                    "type": "number"
                }
            }
        }
    }
}

browser_actions

array

Automate browser interactions before extraction. Supports click, scroll, wait, type, and more.Example:

[
  {"click": {"selector": "#load-more"}},
  {"wait": {"duration": 2000}}
]

network_capture

array

Capture API calls and network requests during page load. Specify URL patterns to intercept.Example:

{
    "url": "https://www.example.com",
    "render": true,
    "network_capture":
    [
        {
            "method": "GET",
            "url":
            {
                "type": "exact",
                "value": "https://www.example.com/api/data"
            }
        }
    ]
}

headers

object

Add custom HTTP headers to your request. Useful for authentication or custom user agents.

Example:

"headers": {
  "User-Agent": "Custom Bot 1.0",
  "X-API-Key": "your-key"
}

array

Set cookies before loading the page. Great for accessing logged-in content or maintaining sessions.Each cookie needs a key, value, and domain.

Example:

"cookies": [
  {
    "key": "session_id",
    "value": "abc123xyz",
    "domain": "example.com"
  }
]

Usage

Basic HTML extraction

Extract HTML content from any URL:

from nimble_python import Nimble

nimble = Nimble(api_key="YOUR-API-KEY")

result = nimble.extract(
    url= "https://www.google.com/search?q=nimble"
)

html = result.data.html
print(html)

JavaScript rendering

Enable rendering for dynamic sites (React, Vue, etc.):

from nimble_python import Nimble

nimble = Nimble(api_key="YOUR-API-KEY")

result = nimble.extract(
    url= "https://www.nimbleway.com/sdk",
    render= True
)

print(result.data.html)

Stealth mode

Bypass anti-bot protections with stealth driver:

from nimble_python import Nimble

nimble = Nimble(api_key="YOUR-API-KEY")

result = nimble.extract(
    url= "https://www.protected-site.com",
    render= True,
    driver= "vx10"
)

print(result.data.html)

Parsing with CSS selectors

Extract structured data with a parsing schema:

from nimble_python import Nimble

nimble = Nimble(api_key="YOUR-API-KEY")

result = nimble.extract(
    url= "https://www.amazon.com/s?k=iphone+17",
    parser= {
        "title": {
            "selector": "h1.product-title"
        },
        "price": {
            "selector": ".price",
            "type": "number"
        },
        "in_stock": {
            "selector": ".availability",
            "type": "boolean"
        }
    }
)

parsed = result.data.parsing
print(f"Title: {parsed['title']}")
print(f"Price: {parsed['price']}")

Browser actions

Automate interactions before extraction:

from nimble_python import Nimble

nimble = Nimble(api_key="YOUR-API-KEY")

result = nimble.extract(
    url="https://www.nimbleway.com/blog",
    render=True,
    browser_actions=[{"wait": 2000}, {"scroll": 500}]
)

print(result.data.html)
print(f"Browser Actions: {result.data.browser_actions}")

Geo-targeting

Extract content from specific locations:

from nimble_python import Nimble

nimble = Nimble(api_key="YOUR-API-KEY")

result = nimble.extract(
    url= "https://ipinfo.io/json",
    country= "GB",
    locale= "en-GB"
)

print(result.data.html)

Drivers

Choose the right extraction engine for your needs:

Driver	Description	Best For	Render
`vx6`	Fast HTTP requests (no JS)	Static HTML, APIs, high volume	No
`vx8`	Headless browser with JS	Dynamic sites, SPAs	Yes
`vx8-pro`	Headful browser with JS	Complex interactions	Yes
`vx10`	Stealth headless browser	Bot-protected sites	Yes
`vx10-pro`	Stealth headful browser	Most protected sites	Yes

Response Fields

Field	Type	Description
`url`	string	The requested URL
`task_id`	string	Unique identifier for the request
`status`	string	`success` or `failed`
`data.html`	string	Extracted HTML content
`data.markdown`	string	Content as markdown (if requested)
`data.parsing`	object	Structured data (if parsing configured)
`status_code`	number	HTTP status code from target

Use cases

High-Scale Extraction

Full control for data extraction at high scale with precise selectors and configurations

Dynamic Content

Handle JavaScript-heavy sites that require full page rendering

Stealth Mode

Bypass anti-bot protections with stealth mode and residential proxies

Data Parsing

Extract structured data using CSS selectors and parsing schemas

Extract vs other tools

What you need	Use
Data from popular sites (Amazon, Google, etc.)	Public Agent - maintained by Nimble
Data from sites not in the gallery	Custom Agent - create with natural language
Data from specific URLs (expert users)	Extract - full control with CSS selectors
Data from entire website	Crawl
Search web + extract content from results	Search

For most users, we recommend starting with Web Search Agents - pre-built extractors maintained by Nimble for popular sites. Use Extract when you need full control over selectors and page interactions.

Features

Explore detailed documentation for each Extract feature:

Async Requests

asynchronous processing and long-running operations

Output Formats

HTML, markdown, and text output options

Geo-Targeting

Extract from specific countries, states, or cities

JS Rendering

Enable JavaScript for dynamic websites

Stealth Mode

Bypass anti-bot systems

Browser Actions

Automate clicks, scrolling, form filling

Parsing Schemas

Define CSS selectors for structured data

Network Capture

Intercept API calls and AJAX requests

Headers & Cookies

Send custom headers and cookies

Next steps

Try Nimble Studio Instead

Create a Web Search Agent for any website — no CSS selectors needed

Install Plugin

Use Nimble in Claude Code or Cursor

API Reference

Explore endpoints, request parameters, and response schemas for the Extract API

Web Search Agents

Learn about intelligent agents that handle extraction automatically

Introduction

Web Search Agents

Web Tools

Guides

SDKs

Admin

​Quick Start

​Example Request

​Example Response

​How it works

​Parameters

​Usage

​Basic HTML extraction

​JavaScript rendering

​Stealth mode

​Parsing with CSS selectors

​Browser actions

​Geo-targeting

​Drivers

​Response Fields

​Use cases

High-Scale Extraction

Dynamic Content

Stealth Mode

Data Parsing

​Extract vs other tools

​Features

Async Requests

Output Formats

Geo-Targeting

JS Rendering

Stealth Mode

Browser Actions

Parsing Schemas

Network Capture

Headers & Cookies

​Next steps

Try Nimble Studio Instead

Install Plugin

API Reference

Web Search Agents

Quick Start

Example Request

Example Response

How it works

Parameters

Usage

Basic HTML extraction

JavaScript rendering

Stealth mode

Parsing with CSS selectors

Browser actions

Geo-targeting

Drivers

Response Fields

Use cases

Extract vs other tools

Features

Next steps