Insurance Premium Calculator
# Insurance Premium Calculator - MCP Server
A simple Model Context Protocol (MCP) server that provides insurance premium calculation services. This server implements three main tools for calculating and comparing insurance premiums.
## Features
### Tools Available
1. **Calculate Basic Premium** - Calculates basic insurance premium based on age, coverage amount, and policy type
2. **Calculate Health Premium with Conditions** - Calculates health insurance premium including pre-existing conditions and smoking status
3. **Compare Policies** - Compares premium for all policy types (term, health, vehicle) for a given age and coverage
## Installation
```bash
npm install
```
## Usage
### Running the Server
To run the MCP server with the inspector:
```bash
npx @modelcontextprotocol/inspector node index.js
```
The server will start and display "Insurance Premium Calculator MCP Server is running!" when ready.
## Tools Documentation
### 1. calculate_basic_premium
Calculates basic insurance premium based on age, coverage and policy type.
**Parameters:**
- `age` (number): Age of the person (18-80)
- `coverage_amount` (number): Coverage amount in rupees
- `policy_type` (enum): Type of insurance policy ("term", "health", "vehicle")
### 2. calculate_health_premium
Calculates health insurance premium including pre-existing conditions.
**Parameters:**
- `age` (number): Age of the person
- `coverage_amount` (number): Coverage amount in rupees
- `pre_existing_condition` (enum): Pre-existing health condition ("none", "diabetes", "hypertension", "heart_disease")
- `smoker` (boolean): Whether the person is a smoker
### 3. compare_policies
Compares premium for all policy types for a given age and coverage.
**Parameters:**
- `age` (number): Age of the person
- `coverage_amount` (number): Coverage amount in rupees
## Technology Stack
- Node.js
- @modelcontextprotocol/sdk - MCP server implementation
- Zod - Data validation
## Testing with Claude Desktop
1. **Ensure you have Claude Desktop installed**
2. **Run the MCP server**:
```bash
npx @modelcontextprotocol/inspector node index.js
```
3. **Open Claude Desktop**
4. **Connect to the MCP server**:
- Claude Desktop should automatically detect the running MCP server
- If not, you may need to follow the instructions in Claude Desktop to manually connect to a local MCP server
5. **Test the tools**:
- Use the Claude Desktop interface to interact with the MCP server
- You can test each tool by providing the required parameters
- Example prompts:
- "Calculate the basic term insurance premium for a 35-year-old with ₹500,000 coverage"
- "What's the health insurance premium for a 40-year-old non-smoker with ₹1,000,000 coverage and diabetes?"
- "Compare all policy types for a 30-year-old with ₹750,000 coverage"
6. **View results**: Claude Desktop will display the JSON responses from the MCP server in a readable format
## Note
Claude Desktop provides an intuitive interface for interacting with MCP servers without having to write custom code. It supports tool discovery, parameter validation, and response visualization.
TDQS
Scored across 3 tools
Each tool has a clearly distinct purpose: calculate_basic_premium handles standard insurance, calculate_health_premium focuses on health-specific factors, and compare_policies provides comparison across types. There is no overlap or ambiguity between these functions.
All tools follow a consistent verb_noun pattern with snake_case (calculate_basic_premium, calculate_health_premium, compare_policies). The naming is predictable and readable throughout the set.
With only 3 tools, the set feels thin for an insurance premium calculator domain. While the tools cover core calculations and comparison, additional operations like updating policies or handling claims might be expected, making the count borderline for the scope.
The tools cover basic premium calculations and comparison, but there are notable gaps for a full insurance domain, such as missing CRUD operations for policies, handling claims, or managing customer data. Agents can work around this for simple tasks but may fail in more complex scenarios.