About MCP Test Client
MCP Test Client is a TypeScript utility for developers testing MCP servers. It connects to servers, calls tools, and validates results, with a mock calculator server included as an example.
Key features
- 01Connect to MCP servers
- 02List available tools
- 03Call named tools
- 04Assert response results
- 05Includes a mock calculator server
Use cases
- 01Test tool calls on an MCP server
- 02Validate tool response results
- 03Learn testing with the calculator example
MCP support
As an MCP client, it connects to servers, lists and calls tools, and checks their responses.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEMCP Test Client
A testing utility for Model Context Protocol (MCP) servers. This client helps you test MCP server implementations by providing a simple interface for making tool calls and validating responses.
Features
- Easy-to-use testing interface for MCP servers
- Built-in support for tool listing and tool calls
- Type-safe implementation using TypeScript
- Assertion utilities for validating server responses
- Mock calculator server implementation for examples
Installation
bun install mcp-test-client
Usage
Basic Example
import { MCPTestClient } from 'mcp-test-client';
describe('MCP Server Tests', () => {
let client: MCPTestClient;
beforeAll(async () => {
client = new MCPTestClient({
serverCommand: 'bun',
serverArgs: ['./path/to/your/server.ts'],
});
await client.init();
});
afterAll(async () => {
await client.cleanup();
});
test('should list available tools', async () => {
const tools = await client.listTools();
expect(tools).toContainEqual(
expect.objectContaining({
name: 'your-tool-name',
description: 'Your tool description',
})
);
});
test('should call a tool', async () => {
await client.assertToolCall(
'your-tool-name',
{ arg1: 'value1', arg2: 'value2' },
(result) => {
expect(result.content[0].text).toBe('expected result');
}
);
});
});
Calculator Server Example
The package includes a mock calculator server for testing and learning purposes:
import { MCPTestClient } from 'mcp-test-client';
describe('Calculator Server Tests', () => {
let client: MCPTestClient;
beforeAll(async () => {
client = new MCPTestClient({
serverCommand: 'bun',
serverArgs: ['./tests/mocks/calculator.ts'],
});
await client.init();
});
afterAll(async () => {
await client.cleanup();
});
test('should perform addition', async () => {
await client.assertToolCall(
'calculate',
{ operation: 'add', a: 5, b: 3 },
(result) => {
expect(result.content[0].text).toBe('8');
}
);
});
});
API Reference
MCPTestClient
Constructor
constructor(config: { serverCommand: string; serverArgs: string[] })
Methods
init(): Promise<void>- Initialize the client and connect to the serverlistTools(): Promise<Tool[]>- Get a list of available tools from the servercallTool(toolName: string, args: Record<string, unknown>): Promise<ToolResult>- Call a specific toolassertToolCall(toolName: string, args: Record<string, unknown>, assertion: (result: ToolResult) => void | Promise<void>): Promise<void>- Call a tool and run assertions on the resultcleanup(): Promise<void>- Clean up resources and disconnect from the server
Development
Prerequisites
- Bun (v1.0.0 or higher)
Setup
- Clone the repository
git clone <repository-url>
cd mcp-test-client
- Install dependencies
bun install
- Run tests
bun test
License
MIT
Contributing
- Fork the repository
- Create your feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add some amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.