Findproxy MCP Documentation
Learn how to integrate the Model Context Protocol server with proxy services for AI-driven proxy management.
Start with Basics
Learn the core concepts of the Model Context Protocol and how it enables AI-driven proxy management.
Read overviewQuick Installation
Get started with Findproxy MCP Server in minutes with our easy-to-follow installation guide.
Install nowAI Integration
Learn how to connect Claude and other AI assistants to your proxy management system.
Explore integrationWhat is MCP?
The Model Context Protocol (MCP) is a standardized communication protocol that allows AI models to request and utilize external contexts and capabilities. Our implementation of MCP focuses on providing proxy management capabilities to AI agents.
With our MCP server implementation, AI agents like Claude can request and manage proxy resources through standardized tools, resources, and prompts defined in the MCP specification. This enables seamless integration with any MCP-compatible client.
Key Benefits
Seamless AI Integration
LLMs can directly access and manage proxy resources
Global Coverage
Access residential IPs from 195+ countries
Security
Managed access with comprehensive auditing
API Extensibility
Extend with custom tools and integrations
Compatible MCP Clients
Findproxy MCP Server works with these MCP-compatible clients:
Architecture
Our MCP implementation follows the architecture described in the official MCP specification, with components designed to provide proxy management capabilities to AI agents while maintaining compatibility with the broader MCP ecosystem.
MCP Architecture Diagram
AI Agents (Claude, GPT-4, etc.)
LLMs using MCP for web access
Proxy MCP Server
TypeScript SDK Implementation
Proxy Management
Purchase, rotation, metrics
Auth & Security
API keys, permissions
Session Management
State persistence
NetNut
Residential proxies
Oxylabs
Future integration
Bright Data
Future integration
MCP Server
Built on the official TypeScript SDK, our server implements the Model Context Protocol specification, handling requests from AI agents and providing tools and resources for proxy management.
Proxy Manager
Manages the lifecycle of proxy connections, including acquisition, rotation, and health monitoring.
Provider Adapters
Standardized interfaces for connecting to various proxy providers like NetNut, with plans to support Oxylabs, Bright Data, and more.
AI Integrations
Specialized adapters for connecting with various LLM platforms, with deep integrations for Claude Desktop.
Installation
Getting started with the Findproxy MCP server is straightforward. You can install it using NPM or run it as a Docker container.
NPM Installation
npm install @modelcontextprotocol/sdk
This installs the official MCP TypeScript SDK which is the foundation for the Findproxy MCP Server.
Docker
docker pull mcpserverdemo/proxy-mcp:latest docker run -p 3000:3000 -e MCP_TRANSPORT=sse mcpserverdemo/proxy-mcp:latest
This pulls and runs our ready-to-use Docker image with all dependencies pre-configured.
Configuration
When fully implemented, the MCP server will be configurable using environment variables or a configuration file, following patterns established in the official MCP TypeScript SDK. Here are the planned configuration options:
| Option | Environment Variable | Description | Default |
|---|---|---|---|
| Server Name | MCP_SERVER_NAME | Name of your MCP server instance | proxy-mcp |
| Port | PORT | Port for HTTP/SSE transport | 3000 |
| Transport | MCP_TRANSPORT | Transport protocol (stdio, sse) | stdio |
| Log Level | LOG_LEVEL | Logging verbosity (debug, info, warn, error) | info |
MCP Tools
The Findproxy MCP Server provides a set of standardized tools that allow AI models to interact with proxy resources. These tools follow the MCP specification and are accessible to any MCP-compatible client.
list-regions
Get available regions for proxies
Lists all available geographic regions where proxies can be provisioned, including countries and cities.
View documentationcreate-proxy
Provision a new proxy
Create a new proxy in a specific geographic region with desired properties and authentication.
View documentationrotate-ip
Change IP address
Request a new IP address for an existing proxy while maintaining the same geolocation parameters.
View documentationExample: Tool Invocation
How to call MCP tools from a client
// Using the MCP client SDK
const result = await mcpClient.callTool({
name: "create-proxy",
arguments: {
country: "Germany",
city: "Berlin",
proxyType: "residential",
dataAllocation: 5
}
});
console.log(result.proxy.connectionString);This example shows how an MCP client can request a new residential proxy in Berlin, Germany with a 5GB data allocation using the standardized create-proxy tool.
list-regions Tool
list-regions
Get available regions for proxies
The list-regions tool retrieves all available geographic regions where proxies can be provisioned. Results can be filtered by country, provider, or proxy type.
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
| country | string | No | Filter regions by country name |
| provider | string | No | Filter regions by provider |
| proxyType | string | No | Filter regions by proxy type (residential, mobile) |
Example Response
{
"regions": [
{
"name": "Germany",
"code": "DE",
"cities": [
{
"name": "Berlin",
"providers": ["Netnut", "Oxylabs"]
},
{
"name": "Munich",
"providers": ["Netnut"]
},
{
"name": "Frankfurt",
"providers": ["Netnut", "Bright Data"]
}
]
},
{
"name": "United States",
"code": "US",
"cities": [
{
"name": "New York",
"providers": ["Netnut", "Bright Data", "Oxylabs"]
},
{
"name": "Los Angeles",
"providers": ["Netnut", "Bright Data"]
}
]
}
]
}Next Steps
Ready to continue exploring Findproxy MCP? Check out these resources: