MCP Client for Testing

What is MCP Client for Testing?

MCP Client for Testing is a minimalistic client for testing tool calls in MCP (Model Context Protocol) servers. It is intended for developers who need to verify tool calls during MCP server development.

How to use MCP Client for Testing?

Install uv, then invoke from the command line using uvx mcp-client-for-testing with a JSON configuration and tool call definition. Alternatively, install the Python package and use the execute_tool function from mcp_client_for_testing.client in your own code. Command line flags allow setting client and server log levels.

Key features of MCP Client for Testing

• Minimalistic design focused on testing tool calls
• CLI invocation via uvx with JSON configuration
• Python API for integration into test suites
• Configurable client and server log levels
• Works with any MCP server that can be launched as a subprocess
• Supports passing environment variables to the server process

Use cases of MCP Client for Testing

• Testing tool calls in a custom MCP server during development
• Verifying tool call parameters, execution, and return values
• Quick smoke tests using an echo MCP server
• Debugging server configuration and environment settings
• Automating tool call verification in CI pipelines

FAQ from MCP Client for Testing

What are the runtime requirements?

The client requires uv to be installed on the system. The server specified in the configuration must also be runnable from the command line.

How do I test a tool call?

Provide a JSON configuration with the server name, command, arguments, and environment, plus a --tool_call JSON specifying the tool name and its arguments.

Can I use it in my own Python code?

Yes. Install the package with uv pip install mcp-client-for-testing and import execute_tool from mcp_client_for_testing.client for async usage.

Is there an example server to test with?

Yes, the echo-mcp-server-for-testing server can be used with uvx to quickly verify the client works.

How do I configure logging?

Use the --client_log_level and --server_log_level flags with standard logging levels such as WARNING and INFO.