# Octoparse MCP for AI Agents AI Agent Connect

> Octoparse lets you turn your AI into a remote control for your web scraping workflows. Instead of jumping between tabs to check crawler progress or manually exporting CSVs, you can just tell your agent to start a task, check the status, or pull the latest data straight into your chat. It handles the heavy lifting of cloud-based extraction so you can focus on the actual analysis.

## Overview
- **Category:** developer-tools
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_YNVzYKGdUzzshKHx6wuFcnKCq4KAE2d7CRRCVwit/ai-agent-connect
- **Tags:** data-extraction, no-code, web-automation, cloud-scraping, structured-data, data-ingestion

## Description

Imagine you're trying to keep up with a competitor's pricing across dozens of different websites every single day. Usually, that means logging into a complex dashboard, starting several separate crawlers, and then waiting for them to finish before you can finally download a file and start the actual work of analysis. It's a fragmented process where you spend more time managing the plumbing of the data than the insights themselves. With this Connector, you skip the middleman entirely. You can just tell your agent to spin up the scrapers you've already built and ask it to summarize the latest results immediately in your chat window. It bridges the gap between your cloud-based scraping infrastructure and your daily workspace. Whether you're grabbing product lists for a new growth campaign or pulling complex datasets for a quarterly report, you can manage the entire lifecycle through a simple conversation. It's one of the core ways Vinkius helps people turn static tools into active participants in their projects. You get to stay in your flow while the agent handles the tedious bits of task management and data retrieval. You no longer have to keep a dozen tabs open just to see if your bots are still running or if they hit a wall. You can just ask for the status and get a clear answer.

## Tools

### stop_task
Stop a running cloud task immediately. Use this to halt a crawler if it's behaving unexpectedly or no longer needed.

### get_task_status
Check the current running status of any cloud task. Use this to see if your bots are still active or finished.

### list_task_groups
List all task groups and folders in your account. Use these IDs to filter which scraping tasks the AI should look at.

### list_tasks
List all configured cloud scraping tasks. This helps the agent see every available project you have set up.

### get_new_data
Get new (non-exported) data from a task

### get_task_data
Export data from a completed scraping task using offset-based pagination. This lets you pull the actual results into your AI's context.

### start_task
Start a cloud scraping task instantly. This tells the engine to begin fetching data from your target websites.

### update_data_status
Mark data as exported

## Prompt Examples

**Prompt:** 
```
Start the Amazon price monitor and tell me when it's done.
```

**Response:** 
```
I've started the 'Amazon Price Monitor' task for you. I'll keep an eye on it, and as soon as it finishes, I'll pull the latest pricing data and summarize any significant changes.
```

**Prompt:** 
```
Get the data from the NYC Real Estate task and make a table.
```

**Response:** 
```
I've pulled the latest rows from the 'Real Estate NYC' task. Here is the summary of the most recent listings:

| Address | SqFt | Price |
| --- | --- | --- |
| 123 Main St | 1,200 | $450,000 |
| 456 Oak Ave | 950 | $390,000 |
```

**Prompt:** 
```
What's the status of my LinkedIn scraper?
```

**Response:** 
```
The 'LinkedIn Profiles Q4' task is currently running. It has successfully captured 4,523 rows of data so far. Would you like me to pull that data for you now?
```

## Capabilities

### Launch and halt scrapers
Start a cloud task or kill a runaway crawler instantly.

### Monitor crawler progress
Check if your bots are running, finished, or stuck in real-time.

### Fetch scraped results
Pull the latest data directly into your conversation for immediate use.

### Manage task folders
Organize and filter through your different scraping projects and groups.

### Update crawler parameters
Change URLs or search keywords without ever opening the dashboard.

### Clear old data
Wipe out testing footprints to keep your production environment clean.

## Use Cases

### Competitive Price Tracking
A business analyst asks the agent to start the 'Daily Retailer' scraper and summarize any price drops over 10% from the latest data.

### Lead Generation
A growth hacker tells the agent to run a LinkedIn scraper and immediately format the results into a CSV-ready markdown table.

### Schema Debugging
A data engineer asks the agent to list all tasks and then pull a sample of the JSON data to verify the fields match the database.

### Dynamic Product Monitoring
An e-commerce owner uses the agent to update the search URL for a product monitor every morning to catch new arrivals.

## Benefits

- Stop manual dashboard checking by using `get_task_status` to see if your scrapers are finished without leaving your workspace.
- Pull data directly into your chat with `get_task_data` so you can start summarizing or formatting results the second they're ready.
- Update your scraping targets on the fly using `update_task_params` to change URLs or keywords without logging into a separate website.
- Keep your workspace clean by using `clear_task_data` to wipe out test results before you start your real production runs.
- Organize your projects easily with `list_task_groups` and `list_tasks` so your agent always knows exactly which scraper to trigger.
- Automate your authentication flow with `get_token` to ensure your agent has consistent, secure access to your scraping account.

## How It Works

The bottom line is you get a conversational interface for your entire cloud scraping pipeline.

1. Subscribe to the Octoparse MCP on the Vinkius marketplace.
2. Provide your Premium Octoparse API credentials (username/email and password).
3. Ask your agent to start a task, check the status, or pull data into your current window.

## Frequently Asked Questions

**Can I use Octoparse MCP to start scrapers automatically?**
Yes, you can tell your agent to trigger any of your existing cloud tasks instantly. This removes the need to manually click buttons in the Octoparse dashboard.

**How do I get the scraped data into my chat?**
The Connector pulls the results directly from Octoparse into your conversation so you can see them immediately. You can then ask the agent to summarize, format, or analyze that data.

**Can I check if my scrapers are still running?**
You can ask the agent for the status of any specific task to see if it's active, finished, or pending. It provides real-time updates on your cloud tasks.

**Does Octoparse MCP work with my current account?**
Yes, it connects to your existing Premium Octoparse account using your API credentials. It works with the projects and tasks you have already set up.

**Can I change the search URL without opening the dashboard?**
Yes, the agent can update your task parameters like URLs or keywords on the fly. This is great for quickly pivoting your scraping targets.

**Is it possible to clear my test data?**
You can ask the agent to purge the data from a specific task to keep your production environment clean. It helps you keep your results organized and accurate.

**Can I have my AI format the scraped JSON into a clean Markdown table?**
Absolutely. Because Octoparse MCP connects natively via the `get_task_data` capability directly into the AI's isolated context window, the language model can instantly translate cumbersome JSON fields into polished, structured, and legible tabular outputs on demand.

**Is it possible to track task progress percentage in the chat?**
Yes. When you instruct your agent to run `get_task_status`, it fetches the real-time runtime progress metrics from Octoparse's cloud. You'll see whether it's Waiting, Running, or Completed, along with how many rows have been extracted so far.

**Do I need a paid Octoparse plan for API capabilities to work?**
Yes. Octoparse explicitly limits their Advanced Cloud APIs strictly to their paid subscription levels. A Free tier account will reject the authentication tokens when attempting to fetch the runtime data.

**Can my AI automatically find the latest extracted data for a specific task?**
Yes! Use the `get_not_exported_data` tool with the Task ID. Your agent will respond with complete metadata for the newest records that haven't been marked as exported yet in seconds.

**How do I find my Octoparse OpenAPI Access Token?**
Log in to Octoparse, navigate to the **OpenAPI** section in your profile or developer portal, and follow the instructions to generate a Bearer token using your account credentials.

**Can I start a scraper via the AI?**
Absolutely. Use the `start_task` tool with your Task ID. The AI will command Octoparse to begin the extraction in the cloud immediately.