test-mcp-server

Automate MCP server testing with e2e, visual regression, live, and eval tests.

210|11|Updated Nov 17, 2025
One-click install
npx skills add https://github.com/Sunpeak-AI/sunpeak --skill test-mcp-server
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: test-mcp-server
Source: https://github.com/Sunpeak-AI/sunpeak/tree/main/skills/test-mcp-server
Command: npx skills add https://github.com/Sunpeak-AI/sunpeak --skill test-mcp-server

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Manage and validate MCP servers across languages with a unified testing framework that covers e2e, visual regression, live testing, and multi-model evals.

Core Features & Use Cases

  • E2E tests using the inspector to validate MCP interactions across hosts.
  • Visual regression tests to ensure UI stability during tool calls.
  • Live tests against real ChatGPT to verify tool behavior in production environments.
  • Evals (multi-model tool calling) to compare tool responses across models.
  • For sunpeak app projects testing integrates automatically; for non-sunpeak MCP servers, sunpeak test init scaffolds a self-contained tests directory.

Quick Start

Run sunpeak test init to scaffold a self-contained MCP testing setup for your project.

Frequently Asked Questions about test-mcp-server

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate end-to-end testing for an MCP server?

Automate MCP server end-to-end testing by scaffolding a test directory with sunpeak test init, which sets up Playwright configurations and fixtures to validate interactions across hosts.

Can I test non-JavaScript MCP servers with Playwright?

Yes, you can test non-JavaScript MCP servers by running sunpeak test init to scaffold a self-contained tests directory, providing Playwright-based tooling to validate various backends.

What is visual regression testing for MCP tool calls?

Visual regression testing for MCP tool calls ensures UI stability during interactions by capturing and comparing visual outputs, preventing unintended interface changes during server execution.

Does MCP testing support live validation against real ChatGPT environments?

Yes, MCP testing supports live validation against real ChatGPT to verify tool behavior in production environments, requiring optional live configurations to enable automated checks.

How do I compare MCP tool responses across multiple AI models?

Compare MCP tool responses across multiple AI models using eval testing, which runs multi-model tool calling to evaluate and contrast tool behaviors against different model backends.

Do I need Playwright configuration to run MCP server tests?

Yes, Playwright configuration is required to run MCP server tests, along with fixtures for the inspector and optional live or eval configurations to enable full automated validation.