> ## Documentation Index
> Fetch the complete documentation index at: https://docs.corrdata.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# MCP Server Overview

> Integrate CorrData with AI assistants using the Model Context Protocol

## What is MCP?

The [Model Context Protocol (MCP)](https://modelcontextprotocol.io) is an open standard that enables AI assistants to securely access external data and tools. CorrData's MCP server allows AI assistants like Claude to:

* Query pipeline assets and measurements
* Analyze risk scores and compliance status
* Generate reports and recommendations
* Access real-time CP monitoring data

<Info>
  CorrData's MCP server is built on Anthropic's official MCP specification, ensuring compatibility with Claude and other MCP-enabled assistants.
</Info>

## Architecture

```
┌─────────────────────────────────────────────────────────────┐
│                     AI Assistant (Claude)                    │
└────────────────────────────┬────────────────────────────────┘
                             │ MCP Protocol
                             ▼
┌─────────────────────────────────────────────────────────────┐
│                    CorrData MCP Server                       │
│  ┌─────────────┐  ┌─────────────┐  ┌─────────────┐         │
│  │   Tools     │  │  Resources  │  │   Prompts   │         │
│  │             │  │             │  │             │         │
│  │ get_asset   │  │ pipelines   │  │ compliance  │         │
│  │ query_      │  │ test_       │  │ risk_       │         │
│  │ measurements│  │ stations    │  │ assessment  │         │
│  │ analyze_    │  │ rectifiers  │  │             │         │
│  │ impact      │  │             │  │             │         │
│  └─────────────┘  └─────────────┘  └─────────────┘         │
└────────────────────────────┬────────────────────────────────┘
                             │ GraphQL
                             ▼
┌─────────────────────────────────────────────────────────────┐
│                     CorrData API                             │
└─────────────────────────────────────────────────────────────┘
```

## Key Benefits

<CardGroup cols={2}>
  <Card title="Natural Language Queries" icon="comments">
    Ask questions like "What's the risk score for the Northern Pipeline?" and get instant answers.
  </Card>

  <Card title="Contextual Recommendations" icon="lightbulb">
    AI can analyze data and provide actionable recommendations based on your pipeline data.
  </Card>

  <Card title="Compliance Assistance" icon="shield-check">
    Get help understanding PHMSA requirements and identifying compliance gaps.
  </Card>

  <Card title="Secure Access" icon="lock">
    All data access is authenticated and scoped to your organization's permissions.
  </Card>
</CardGroup>

## Quick Start

### 1. Install the MCP Server

```bash theme={null}
pip install corrdata-mcp
# or
uvx corrdata-mcp
```

### 2. Configure Claude Desktop

Add to your `claude_desktop_config.json`:

```json theme={null}
{
  "mcpServers": {
    "corrdata": {
      "command": "uvx",
      "args": ["corrdata-mcp"],
      "env": {
        "CORRDATA_API_URL": "https://api.corrdata.com/graphql",
        "CORRDATA_API_KEY": "your-api-key"
      }
    }
  }
}
```

### 3. Start Using

Once configured, you can ask Claude questions like:

* "Show me all rectifiers with below-threshold output"
* "What's the compliance status for the Eastern Region?"
* "Analyze the risk factors for segment S-1234"
* "Generate a monthly CP summary report"

## Next Steps

<CardGroup cols={2}>
  <Card title="Available Tools" icon="wrench" href="/mcp/tools">
    Explore all MCP tools
  </Card>

  <Card title="Configuration" icon="gear" href="/mcp/configuration">
    Advanced configuration options
  </Card>

  <Card title="Examples" icon="code" href="/mcp/examples">
    Usage examples and recipes
  </Card>
</CardGroup>