MCP Test Server

A comprehensive Model Context Protocol (MCP) server designed specifically for testing MCP scanners and validating MCP implementations. This server provides a rich set of tools, resources, and prompts to exercise various aspects of the MCP specification.

🔍 Scanner Compatible: This repository is fully configured to be detected by the APIsec MCP Audit Scanner.

⚡ Quick Start: Want to test scanner detection immediately? See QUICK_START.md

Scanner Detection Documentation

QUICK_START.md - Get scanned in 10 minutes
SCANNER_GUIDE.md - Detailed detection guide
SCANNER_CHECKLIST.md - Verification checklist

Features

🛠️ Tools

The server exposes multiple tools with varying complexity levels:

echo - Basic string echo for testing simple parameter handling
add_numbers - Numeric operations testing
format_json - JSON object handling and formatting
list_operations - Array/list manipulation (sort, reverse, count, join)
complex_schema - Nested object schemas with various types
timestamp - Tools with optional parameters only

📦 Resources

Multiple resource types for testing resource discovery and reading:

static-text - Plain text resource
json-data - Structured JSON data
markdown-doc - Formatted markdown documentation
config - Configuration file example

💬 Prompts

Sample prompts to test prompt capabilities:

test-prompt - Basic prompt with required argument
debug-prompt - Multi-argument prompt with optional parameters

Installation

From Source

# Clone the repository git clone <repository-url> cd mcp-test # Install the package pip install .

Development Installation

# Install with development dependencies pip install -e ".[dev]"

Usage

Running the Server

Command Line

After installation, run the server directly:

mcp-test-server

Python Module

Alternatively, run as a Python module:

python server.py

Testing with MCP Client

The server uses stdio transport, so you can test it with any MCP-compatible client:

import asyncio from mcp import ClientSession, StdioServerParameters from mcp.client.stdio import stdio_client async def test_mcp_server(): server_params = StdioServerParameters( command="mcp-test-server", args=[], ) async with stdio_client(server_params) as (read, write): async with ClientSession(read, write) as session: await session.initialize() # List available tools tools = await session.list_tools() print(f"Available tools: {[t.name for t in tools.tools]}") # Call a tool result = await session.call_tool("echo", {"message": "Hello MCP!"}) print(f"Result: {result}") if __name__ == "__main__": asyncio.run(test_mcp_server())

Configuration for MCP Clients

Add to your MCP client configuration (e.g., Claude Desktop config):

{ "mcpServers": { "mcp-test": { "command": "mcp-test-server" } } }

Or with explicit Python path:

{ "mcpServers": { "mcp-test": { "command": "python", "args": ["-m", "server"] } } }

Testing MCP Scanners

This server is ideal for testing MCP scanner tools because it provides:

Diverse Tool Schemas - From simple strings to complex nested objects
Multiple Resource Types - Different MIME types and content structures
Edge Cases - Tools with no required parameters, optional fields, enums
Standard Compliance - Follows MCP specification strictly
Scanner Detection - Multiple configuration files for detection testing

Scanner Detection

This repository is configured to be detected by MCP scanner tools like the APIsec MCP Audit Scanner.

Detection files included:

mcp.json - Claude Desktop style configuration
mcp.yaml - YAML format configuration
.mcp/config.json - Hidden directory config
package.json - npm dependencies with @modelcontextprotocol/sdk
requirements.txt - Python dependencies with modelcontextprotocol

📖 See for detailed scanner testing instructions.

Scanner Test Checklist

Discovers all 6 tools
Parses complex nested schemas correctly
Identifies all 4 resources with correct URIs
Handles tools with optional-only parameters
Recognizes prompt capabilities
Correctly interprets enum constraints
Handles array and object types

Project Structure

mcp-test/ ├── server.py # Main MCP server implementation ├── pyproject.toml # Package configuration ├── package.json # npm metadata (for scanner detection) ├── requirements.txt # Python dependencies ├── README.md # This file ├── SCANNER_GUIDE.md # Scanner detection guide ├── .gitignore # Git ignore rules ├── LICENSE # MIT License ├── mcp.json # MCP configuration (Claude Desktop style) ├── mcp.yaml # MCP configuration (YAML format) ├── mcp-config.json # Example client configuration ├── .mcp/ # MCP metadata directory │ ├── config.json # Scanner-detectable config │ └── mcp.json # MCP metadata ├── examples/ # Example usage scripts │ ├── README.md │ ├── test_client.py │ └── scanner_test.py └── tests/ # Unit tests ├── __init__.py └── test_server.py

Requirements

Python 3.10 or higher
mcp >= 0.9.0

Development

Running Tests

pytest tests/

Code Style

This project follows PEP 8 guidelines. Format code with:

black server.py

API Reference

Tools

echo

Input: message (string, required)
Output: Echoes the input message
Purpose: Test basic string parameter handling

add_numbers

Input: a (number), b (number)
Output: Sum of the two numbers
Purpose: Test numeric parameter handling

format_json

Input: data (object), indent (number, default: 2)
Output: Formatted JSON string
Purpose: Test object parameter handling

list_operations

Input: items (array), operation (enum), separator (string, optional)
Output: Result of list operation
Purpose: Test array handling and enum constraints

complex_schema

Input: Nested object with user info and options
Output: Processed data structure
Purpose: Test complex nested schema parsing

timestamp

Input: format (enum, optional)
Output: Current timestamp in specified format
Purpose: Test tools with all-optional parameters

Resources

All resources use URIs in the format mcp://test/{resource-name}:

mcp://test/static-text - Plain text content
mcp://test/json-data - Structured JSON
mcp://test/markdown-doc - Markdown documentation
mcp://test/config - Configuration data

Contributing

Contributions are welcome! Please:

Fork the repository
Create a feature branch
Make your changes
Add tests if applicable
Submit a pull request

License

MIT License - see LICENSE file for details

Support

For issues, questions, or contributions, please open an issue on the repository.

Changelog

v0.1.0 (Initial Release)

Basic MCP server implementation
6 diverse tools for testing
4 resource types
2 sample prompts
Complete documentation

Note: This is a testing server. Do not use in production environments.

MCP Test Server

Scanner Detection Documentation

Features

🛠️ Tools

📦 Resources

💬 Prompts

Installation

From Source

Development Installation

Usage

Running the Server

Command Line

Python Module

Testing with MCP Client

Configuration for MCP Clients

Testing MCP Scanners

Scanner Detection

Scanner Test Checklist

Project Structure

Requirements

Development

Running Tests

Code Style

API Reference

Tools

echo

add_numbers

format_json

list_operations

complex_schema

timestamp

Resources

Contributing

License

Support

Changelog

v0.1.0 (Initial Release)

Related Projects

Resources

New MCP Servers

Latest Blog Posts

MCP directory API