# ParseHub MCP for AI Agents AI Agent Connect

> ParseHub MCP lets you control cloud scraping projects directly from your AI client. You can launch headless browser runs, check progress in real-time, and pull down structured JSON datasets without leaving your chat window. It turns complex web extraction into a simple conversation.

## Overview
- **Category:** developer-tools
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_KQ4Vz5JSUvjaARvQvBDyCn0dVVNm7RhifigMEXcR/ai-agent-connect
- **Tags:** data-extraction, headless-browser, web-crawling, json-output, cloud-scraping, automation-workflows

## Description

Imagine you're trying to track competitor pricing across hundreds of pages. Usually, this means jumping between a dashboard, checking run statuses, and manually downloading files. With this Connector, you just tell your AI client to start the scrape. It handles the heavy lifting of headless browser automation in the cloud. You can see exactly where a job sits in the queue or if it's finished, and then grab the final data for your spreadsheet or database. It's like having a data scraping team that works on command. Because it's part of the Vinkius catalog, you can plug it into your existing workflow and stop worrying about the plumbing of web extraction. You focus on the insights; the Connector handles the scraping.

## Tools

### get_project
Shows the configuration details for a specific scraping project. It helps you see the start URLs and templates.

### list_projects
Lists all the web scraping projects currently in your account. It provides the tokens needed for all other operations.

### get_run_data
Downloads the final JSON payload from a completed run. This gives your agent the actual data it needs to process.

### get_run_details
Shows the current status of a run, like whether it's queued or running. Use this to monitor progress in real time.

### get_last_ready_data
Grabs the most recent completed data for a project without needing a run token. It is perfect for dashboard updates.

### cancel_run
Stops a running or queued scrape to free up your queue slots. This is useful for stopping jobs that are no longer needed.

### delete_run
Permanently removes old runs from your account. Use this to clear out history and free up storage quota.

### list_runs
Useful for auditing or finding a specific completed run to fetch data from.

Get the history of all runs for a ParseHub project

### run_project
Starts a new scraping run for a specific project. This puts the job into the cloud queue for processing.

### run_project_with_url
Starts a run on a custom URL while using a project's template. It lets you scrape different pages with the same rules.

## Prompt Examples

**Prompt:** 
```
Show me all the scraping projects I have.
```

**Response:** 
```
### Your ParseHub Projects

| Project Name | Token | Last Run |
| :--- | :--- | :--- |
| **Pricing Monitor** | `t9zx...` | 2026-07-10 |
| **Real Estate Leads** | `t4ax...` | 2026-07-08 |
| **Competitor Blogs** | `t1bb...` | 2026-07-05 |

Would you like me to launch one of these?
```

**Prompt:** 
```
Run the pricing project and tell me if it's done.
```

**Response:** 
```
Dispatched action for target `t9zx...`. 

**Current Status:** 
- **Run ID:** `run_k1l` 
- **Status:** `queued` 

The parser will begin extracting domains soon. I'll let you know when it's finished.
```

**Prompt:** 
```
Give me the latest data from the pricing project.
```

**Response:** 
```
Pulled the latest data from your project. The JSON contains **40 records** with the following fields:

- **Product Name**
- **Current Price**
- **Discount %**
- **Store URL**

Would you like me to format this into a table or save it to a file?
```

## Capabilities

### Launch cloud scraping jobs
Start a new headless browser run for a specific project or custom URL.

### Track real-time progress
Check if a scrape is queued, running, or complete without leaving your chat.

### Retrieve structured JSON payloads
Pull down the final scraped data as a clean JSON array for your agent to process.

### List scraping projects
See all your pre-configured scraping targets and their recent history.

### Stop long-running scrapes
Cancel active jobs instantly to free up your queue slots.

### Fetch latest dashboard data
Grab the most recent completed data without needing to trigger a new run.

### Clear old run history
Permanently delete old runs to manage your account storage quota.

## Use Cases

### Competitor Price Tracking
A marketing lead asks the agent to run the 'Pricing Monitor' project and fetch the latest JSON to compare costs.

### Real Estate Lead Gen
A user tells the agent to scrape a specific new listing page using the 'Real Estate' template to grab details.

### Academic Research
An analyst kicks off a scraper for a set of journals and asks the agent to summarize the results once finished.

### Data Pipeline Automation
An engineer uses the agent to trigger a daily scrape and then asks it to format the data for a CSV file.

## Benefits

- Stop manual dashboard checking by using get_run_details to see status updates in your chat window.
- Save storage space by using delete_run to wipe out old data logs and keep your account clean.
- Get data faster by using get_last_ready_data for real-time dashboard updates without extra steps.
- Run custom scrapes on specific pages using run_project_with_url without rebuilding your templates.
- Manage multiple scraping tasks at once with list_projects to see your full portfolio in one view.
- Control your queue better by using cancel_run to stop jobs that aren't needed anymore to save resources.

## How It Works

The bottom line is you get to run complex web scraping via chat instead of manual dashboard management.

1. Connect your ParseHub API key to the Connector.
2. Tell your AI client which project to run or what URL to scrape.
3. Receive the final structured data or a status update.

## Frequently Asked Questions

**Can ParseHub MCP run headless browsers for me?**
Yes, it triggers cloud-based headless runs. This means you don't have to worry about browser compatibility, driver updates, or IP blocking on your own machine.

**How does ParseHub MCP handle large datasets?**
It pulls down structured JSON payloads. Your agent can then take that data and summarize, filter, or format it however you need.

**Can I scrape a specific URL instead of the default project URL?**
Yes, you can use the custom URL tool. This allows you to apply your project's existing scraping rules to a unique page or a different category.

**Does ParseHub MCP save my account storage?**
You can use the delete tool to remove old runs. This helps you keep your account storage free and ensures you don't hit your quota limits.

**Is it good for real-time dashboard updates?**
Yes, the 'get last ready data' tool is perfect for this. It pulls the most recent results without needing to start a new run every time you want an update.

**How do I know when a scrape is finished?**
Your agent can check the status of a run automatically. It will notify you as soon as the status moves to 'complete' so you can grab the data immediately.

**Do I need the ParseHub Desktop tool running to use this?**
No. This integration operates completely natively via ParseHub's Cloud API endpoints. You only need the desktop app to build the templates originally. All executions mapped here happen on their cloud scaling servers.

**Can I provide a different Start URL when running a project?**
Yes. The `run_project_with_url` command allows you to explicitly provide a `start_url` query property. This instructs the ParseHub crawler to ignore its project-saved URL and begin parsing the newly mapped domain using the same semantic template.

**Is the downloaded data returned in JSON or raw HTML?**
The payload fetched by `get_run_data` is exported entirely as structured, pre-parsed JSON mirroring the exact template node selections defined in your project architecture.