# Browserhub MCP for AI Agents MCP

> Browserhub provides AI agents with immediate access to enterprise web scraping capabilities. Your agent can run structured scrapers or perform one-off URL extractions using real browsers. It manages job monitoring, tracks task history, and controls proxy rotation so you get reliable, clean data directly into your workflow.

## Overview
- **Category:** data-management
- **Price:** Free
- **Tags:** proxy-management, anti-bot, javascript-rendering, data-extraction, web-scraping, automation

## Description

Connecting Browserhub lets your AI client manage complex web scraping tasks through simple conversation. Instead of writing boilerplate code for every website structure, your agent simply tells the MCP what to extract from a given URL or list of URLs. It handles everything needed—from managing proxy locations and checking account credits to launching and monitoring multi-step scraping jobs.

If you're working with large datasets, this MCP ensures your data extraction remains reliable, even if sites change their layout. You can instruct your agent to check which scrapers are already set up or trigger a brand new job right out of the box. Because Vinkius hosts all these connections in one place, you get access to industrial-strength web automation without needing multiple specialized services. The result is clean, structured data waiting for you to analyze.

## Tools

### direct_scrape
Runs a quick scrape on a single URL without needing to create or select a specific scraper blueprint first.

### get_account_balance
Checks how many credits are remaining in your Browserhub account.

### get_blueprint
Retrieves specific details and metadata for a single, defined scraping blueprint.

### get_scraping_job
Gets the status and final results of a previously run or currently running job.

### get_scraper
Retrieves details about one specific, configured scraper setup.

### list_blueprints
Shows a full list of every scraping blueprint you have set up in your account.

### list_scraping_jobs
Lists the history and metadata for all past web scraping jobs.

### list_proxy_locations
Shows every available proxy location you can use to prevent IP blocking during scraping.

### list_scrapers
Lists all scrapers and their configurations that you have defined in your account.

### run_scraper
Starts a new scraping job using one of your existing, named scraper blueprints.

## Prompt Examples

**Prompt:** 
```
Check my account balance and list all my configured scrapers.
```

**Response:** 
```
**Account Status:**
*Credits Remaining:* 1,500
*Estimated Runs Left:* ~200

**Active Scrapers Found:**
1. **E-commerce Prices (ID: scr_1):** Tracks competitor pricing from the top 3 retail sites.
2. **News Articles (ID: scr_2):** Captures headlines and authors from major news outlets.
3. **Lead Generator (ID: scr_3):** Scrapes contact details from professional networking profiles.
```

**Prompt:** 
```
Scrape the URL for the new quarterly report I just found.
```

**Response:** 
```
✅ Scraping job `job_99283` started successfully for the specified URL.

I've initiated a real browser scrape. This takes time, but I'll monitor it and notify you immediately when the structured data—including tables and contact info—is ready to pull into your workspace.
```

**Prompt:** 
```
What are my scraping jobs history for last month?
```

**Response:** 
```
**Job History Summary:**
* **job_98123 (Date: 6/10):** Status: Complete. Result: 4,500 records extracted.
* **job_97112 (Date: 5/28):** Status: Failed. Error: Timeout on specific domain.
* **job_96001 (Date: 5/1):** Status: Complete. Result: 1,200 records extracted.
```

## Capabilities

### Get Account Credit Balance
Checks your remaining account credit balance to ensure you have enough funds for scraping jobs.

### List All Scrapers and Blueprints
Retrieves comprehensive lists of all scrapers and blueprints configured in the system.

### Run a Defined Scraper Job
Starts a scraping job using an existing, named scraper blueprint for consistent data retrieval.

### Directly Scrape a URL
Performs immediate extraction on a single web page without needing to select or configure a specific scraper first.

### Manage Proxy Locations
Lists all available proxy locations, which helps prevent IP bans during large-scale scraping operations.

## Use Cases

### Tracking Competitor Pricing Changes
A market researcher needs to know if five key competitors changed their pricing on specific product pages. They tell their agent, 'Use the E-commerce scraper for the latest prices.' The agent runs the job and returns a clean spreadsheet of current vs. previous data.

### Building Training Datasets
A data scientist needs thousands of structured articles on a specific topic. They ask their agent to list all available scrapers, pick one for 'News Articles,' and run it across 100 different domains using proxy rotation.

### Real-time Content Monitoring
A developer wants to check if a key piece of content was updated on a partner site. Instead of building a whole scraper, they use the agent and ask it to `direct_scrape` that single URL, getting immediate validation.

### Auditing Scraper Health
A developer suspects one of their existing scrapers is broken or outdated. They tell the agent to `get_scraper` details for a specific blueprint and check its metadata instantly before attempting a job run.

## Benefits

- Monitor ongoing jobs with the `get_scraping_job` tool. You always know if your scraping task is running, stalled, or ready for data retrieval.
- Don't waste time setting up scrapers from scratch. Use `list_scrapers` and `list_blueprints` to see everything you already have configured right away.
- Need a quick look at one page? The `direct_scrape` tool lets your agent pull data immediately without needing pre-defined automation rules.
- Stay safe from IP bans. Use the proxy management tools, like `list_proxy_locations`, to ensure every scraping session uses fresh, available IPs.
- Always know your budget. Run `get_account_balance` whenever you start a large project so your agent doesn't hit an unexpected credit limit mid-run.

## How It Works

The bottom line is you get automated web data collection without writing custom API calls or managing proxy lists manually.

1. First, connect your AI client to this MCP using your Browserhub API Key.
2. Next, instruct your agent on the task—for example, 'Scrape these 5 URLs' or 'Run my E-commerce scraper for today's prices.'
3. The system executes the job and returns structured data, which your agent can then process directly into your workspace.

## Frequently Asked Questions

**How does the Browserhub MCP help me scrape multiple websites?**
It manages the complexity of web scraping by handling proxy rotation and IP management. You don't worry about being detected or blocked; the agent handles switching IPs automatically to keep your jobs running smoothly.

**Can I use Browserhub MCP if I don't have a specific scraper set up?**
Yes, you can. The direct scraping capability lets your agent perform one-off extractions on any URL immediately, even if you haven't created a complex blueprint for it.

**Is this MCP better than writing custom Python web scrapers?**
For most people, yes. This MCP lets your agent manage the scraping process via natural language prompts, saving you hundreds of lines of code and maintenance time when websites change their structure.

**What if I need to check my current scrape job status?**
You simply ask your agent to check the job by referencing its ID. It will give you a real-time update on whether it's running, waiting for proxies, or fully finished.

**Can I list all my available scrapers using Browserhub MCP?**
Absolutely. The agent can list every scraper and blueprint you have configured in your account, giving you a quick overview of all your data collection assets.