# Decodo MCP for AI Agents AI Agent Connect

> Decodo (Smartproxy) MCP lets you manage your proxy network directly through your AI agent. You can list endpoints, monitor real-time bandwidth, and audit scraper stats without leaving your workspace. It connects your Smartproxy account to your agent so you can handle infrastructure tasks via natural language.

## Overview
- **Category:** developer-tools
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_xB2XsUTDzl39iALACVI4U1l9wBhj5vaW3vjEGV5S/ai-agent-connect
- **Tags:** proxy-network, web-scraping, network-infrastructure, ip-management, bandwidth-monitoring

## Description

This Decodo (Smartproxy) MCP connects your proxy network infrastructure to your AI agent. Managing a proxy network usually means constant tab-switching between your terminal and a dashboard. You have to copy-paste IP pools, check if you're about to hit a data cap, and hunt down specific sub-user metrics across multiple pages. This Connector changes that by bringing your account data directly into your chat interface. Instead of hunting for a specific port or checking a bandwidth graph, you just ask your agent to show you what's happening. You can pull active residential configs, audit your scraper success rates, or verify your subscription limits in one go. It’s a lot faster for anyone who needs to keep their scraping pipelines running without the constant manual overhead. If you're looking for a way to keep your network management tight, this is a solid addition to your Vinkius setup. You get a direct line to your account data, making it way easier to manage large-scale scraping operations or multi-user proxy setups.

## Tools

### list_datacenter_proxies
Lists your datacenter proxy configurations for high-speed tasks. Use this to see what's active in your data center pool.

### list_proxy_endpoints
Shows all available proxy endpoints currently in your pool. This is great for getting fresh hostnames for your scripts.

### list_residential_proxies
Fetches all residential proxy configurations for your scraping needs. It gives you the specific details you need for high-quality traffic.

### list_sub_users
Lists every nested sub-user tied to your main account. Use this to keep track of your team's access levels.

### get_scraper_stats
Retrieves success rates and hit counts for your scraping APIs. This helps you spot failures in your scraping pipeline immediately.

### list_dedicated_proxies
Shows your assigned dedicated proxy networks for private use. It helps you verify your specific network assignments.

### list_subscriptions
Provides a full list of your active proxy subscriptions. Use this to audit your plans and active seats.

### get_account_details
Pulls your general account health and profile information. Use this to verify your account status quickly.

### get_traffic_usage
Shows your current bandwidth consumption and remaining data cap. This prevents you from running out of data mid-job.

## Prompt Examples

**Prompt:** 
```
Show me how much bandwidth I have left for this month.
```

**Response:** 
```
Checking your current usage...

| Metric | Value |
| :--- | :--- |
| **Total Cap** | 100 GB |
| **Used** | 78.4 GB |
| **Remaining** | 21.6 GB |
| **Status** | 78% Consumed

You have plenty of room left for your current scraping tasks.
```

**Prompt:** 
```
List all my active datacenter proxies.
```

**Response:** 
```
Here are your active Datacenter Proxy Endpoints:

* **US-East Pool**: `gate.dc.smartproxy.com:20000`
* **EU-Central Pool**: `gate.dc.smartproxy.com:20001`

Both are currently operational.
```

**Prompt:** 
```
How is our scraper performing on the eCommerce site?
```

**Response:** 
```
I've pulled the latest stats for your scraping APIs:

* **Total Hits**: 4,025
* **Success Rate**: 96.9%
* **Blocked/Timeout**: 124

The success rate is currently very healthy.
```

## Capabilities

### Pull proxy lists
Get a clear view of your active datacenter and residential endpoints.

### Track bandwidth
See exactly how much data you've used against your monthly cap.

### Audit scraper performance
Check success rates and hit counts for your scraping APIs.

### Manage sub-users
View and organize nested users for your proxy accounts.

### Verify subscriptions
Get a quick summary of your active plans and account health.

## Use Cases

### Debugging a failing scraper
A developer asks the agent to check get_scraper_stats to see if the IP pool is being blocked or just timing out.

### Provisioning new crawler nodes
A network admin uses list_residential_proxies to grab a fresh set of configurations for a new scraping cluster.

### Managing client accounts
A SaaS operator uses list_sub_users to verify that a new client has the correct proxy access.

### Pre-flight bandwidth check
A data engineer asks the agent to check get_traffic_usage before launching a heavy migration task.

## Benefits

- Stop manual copy-pasting of proxy endpoints by using list_proxy_endpoints to feed your scripts directly. This keeps your development flow moving without constant tab switching.
- Avoid hitting data caps early by checking real-time limits with get_traffic_usage before starting big jobs. You'll always know exactly how much bandwidth you have left.
- Identify failing scrapers faster by pulling success rates from get_scraper_stats to spot issues immediately. You can see exactly which APIs are hitting walls in seconds.
- Organize large teams without the friction of dashboard navigation by using list_sub_users to manage permissions. It makes auditing your account structure much faster.
- Audit your infrastructure quickly by using list_datacenter_proxies to see exactly what's active. You can verify your pool configuration without leaving your IDE.

## How It Works

The bottom line is you stop clicking through dashboards and start managing your proxy network with plain English.

1. Subscribe to the Decodo MCP on Vinkius.
2. Input your Smartproxy API key into your agent's settings.
3. Ask your agent to list proxies, check traffic usage, or pull scraper stats.

## Frequently Asked Questions

**What can the Decodo MCP do for my scraping tasks?**
It lets you query your proxy network directly. You can pull endpoint configurations, check your bandwidth usage, and see real-time success rates for your scraping APIs without leaving your workspace.

**Can I check my proxy bandwidth with Decodo MCP?**
Yes. You can ask your agent to check your traffic usage at any time. It will tell you how much data you've used against your monthly cap so you don't run out mid-job.

**How do I manage sub-users using Decodo MCP?**
You can use the Connector to list all nested sub-users associated with your account. This makes it much easier to audit permissions and see who has access to your proxy pools.

**Does Decodo MCP work with Cursor or Claude?**
Yes, it works with any MCP-compatible client. Once connected via Vinkius, you can use it in Claude, Cursor, Windsurf, and other supported environments.

**Can I see my residential proxy configs?**
Yes, the Connector can pull your residential proxy configurations instantly. This is helpful for getting the exact details you need for high-quality web scraping.

**How does Decodo MCP help with rate limits?**
It allows you to quickly list your dedicated proxy assignments. This helps you verify which networks are assigned to your account to help troubleshoot rate-limiting issues.

**Can the AI generate a script with my actual proxy endpoints already inserted?**
Yes. While prompting inside an IDE like Cursor, you can ask 'Write a Python script using my residential proxy endpoint.' The AI agent will call `list_residential_proxies` in the background, harvest the legit port numbers, and assemble runnable code on the spot.

**Is it safe to pull network architecture metrics inside Claude AI?**
Absolutely secure. The Connector protocol bridges interactions locally. At no setup phase does Vinkius cloud servers retrieve your Smartproxy metrics directly—the exchange passes via your local agent bridging right into Decodo's endpoint.

**Can it tell me my remaining bandwidth limit?**
Yes. Ask the bot 'How much target data do I have left?' and it will orchestrate the results of both `get_account_details` (which returns quota details) and `get_traffic_usage` (which exposes used data) to compute the remaining bytes in conversational form.