MCP Server Overview

MCP Server Overview

1. What is MCP?

MCP (Model Context Protocol) is an open protocol developed by Anthropic for connecting AI assistants with external data sources and tools. RAILWISE MCP Server allows AI assistants to access monitoring data, execute commands, and generate reports.

2. Architecture

┌─────────────┐     ┌─────────────┐     ┌─────────────┐
│   AI Client  │ ←→ │  MCP Server  │ ←→ │  RailWise API │
│  (Claude/   │     │  (RailWise)  │     │  (Backend)   │
│   Cursor)   │     │              │     │              │
└─────────────┘     └─────────────┘     └─────────────┘

3. Capabilities

3.1 Tools

Tool Name	Description	Parameters
query_data	Query monitoring data	project_id, point_id, date_range
generate_report	Generate monitoring report	project_id, report_type, date_range
analyze_trend	Analyze data trend	project_id, point_id, period
check_alert	Check warning status	project_id, level
export_data	Export monitoring data	project_id, format, date_range

3.2 Resources

Resource	Description	Access Method
project://{id}	Project information	Read
data://{project_id}/{point_id}	Monitoring data	Read
report://{project_id}/{report_id}	Monitoring report	Read
standard://{code}	Standard document	Read

4. Integration Methods

4.1 Claude Desktop Integration

1. Install RAILWISE MCP Server
2. Configure Claude Desktop settings
3. Add MCP server configuration

{
  "mcpServers": {
    "railwise": {
      "command": "npx",
      "args": ["@railwise/mcp-server"],
      "env": {
        "RAILWISE_API_KEY": "your_api_key"
      }
    }
  }
}

4.2 Cursor Integration

1. Open Cursor Settings
2. Go to Features > MCP
3. Add MCP server

4.3 Custom Client Integration

import { Client } from "@modelcontextprotocol/sdk/client/index.js";

const client = new Client({
  name: "my-client",
  version: "1.0.0"
});

await client.connect(transport);

// List available tools
const tools = await client.listTools();

// Call tool
const result = await client.callTool({
  name: "query_data",
  arguments: {
    project_id: "proj-001",
    point_id: "JC-01",
    date_range: "2024-01-01,2024-01-31"
  }
});

5. Use Cases

5.1 Natural Language Data Query

User: "Query the settlement data of point JC-01 in the last 7 days"
AI: Calls query_data tool → Returns data → Displays results

5.2 Intelligent Report Generation

User: "Generate a weekly report for project proj-001"
AI: Calls generate_report tool → Returns report → Displays report

5.3 Trend Analysis

User: "Analyze the settlement trend of point JC-01"
AI: Calls analyze_trend tool → Returns analysis → Displays trend chart

6. Security

6.1 Authentication

MCP Server uses API Key authentication. All requests must include a valid API Key.

6.2 Permission Control

MCP Server inherits the permissions of the API Key. Only authorized data and operations can be accessed.

6.3 Data Privacy

• MCP Server does not store conversation content
• All data access is through the RailWise API
• Data transmission uses HTTPS encryption

7. Related Documents

• MCP Server Installation
• MCP Server Integration
• MCP Server Tools
• MCP Server Security

---

RailWise · Smart Monitoring — Engineering Surveying and Safety Monitoring Digital Solutions