skills.rest
umbraco
umbracoOFFICIAL
Explore all skills by this author →
WebsiteGitHub

mcp-testing

Define and execute LLM-based acceptance tests for MCP tools.

2|3|Updated Jan 15, 2026
Search Tags:#testing#devops#llm#mcp#acceptance tests#workflow verification
One-click install
npx skills add https://github.com/umbraco/Umbraco-MCP-Base --skill mcp-testing-umbraco
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: mcp-testing
Source: https://github.com/umbraco/Umbraco-MCP-Base/tree/main/plugins/skills/mcp-testing
Command: npx skills add https://github.com/umbraco/Umbraco-MCP-Base --skill mcp-testing-umbraco

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @umbraco-cms/mcp-server-sdk/evals, @jest/globals, ts-jest, and includes references (resource) components.

What problem does it solve?

This Skill streamlines the process of writing and running LLM-based acceptance tests for your MCP tools, ensuring they function correctly within AI-driven workflows.

Core Features & Use Cases

  • LLM-based Acceptance Testing: Verify that your MCP tools behave as expected when controlled by an LLM.
  • Scenario Testing: Define complex multi-step workflows for the LLM to execute.
  • Use Case: You've built a new MCP tool for managing Umbraco content. Use this Skill to write an eval test that instructs an LLM to create, update, and then delete a content item, confirming the tool's end-to-end functionality.

Quick Start

Use the mcp-testing skill to run an eval test for the 'entity-crud' scenario.

Frequently Asked Questions about mcp-testing

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

FAQPage Schema
How do I write LLM-driven acceptance tests for MCP tools?▼

You can write LLM-driven acceptance tests by defining complex multi-step scenarios that an LLM executes against your MCP tools. This framework facilitates scripting these interactions to verify end-to-end workflow functionality and tool integration.

What is the best way to verify MCP tool integration in AI workflows?▼

Verifying MCP tool integration in AI workflows is achieved by running scenario-based evals that instruct an LLM to execute multi-step actions. This confirms the tool behaves as expected within the broader AI-driven process.

Do I need @umbraco-cms/mcp-server-sdk/evals to run MCP testing scenarios?▼

Yes, you need @umbraco-cms/mcp-server-sdk/evals to execute and configure the LLM-based acceptance testing scenarios. The testing framework relies on this SDK dependency to facilitate test runs.

Can I test multi-step workflows with an LLM using this testing framework?▼

Yes, you can test multi-step workflows by defining complex scenarios for the LLM to execute. You can verify end-to-end MCP tool functionality by instructing the LLM to perform sequential actions like creating, updating, and deleting content.

How does scenario testing debug LLM interactions with MCP tools?▼

Scenario testing debugs LLM interactions by executing defined multi-step workflows and observing the LLM's control over MCP tools. This process highlights integration failures and unexpected behavior within AI-driven workflows.

Related Skills

Explore More Software Engineering Skills→
yuhle-qugyuhle-qug

jest-testing-expert

Design, configure, and execute Jest test suites for JavaScript and TypeScript projects.

Community
Advanced
Miosa-osaMiosa-osa
226

error-analysis

Analyze LLM execution traces to categorize failures and generate an error taxonomy.

Official
Advanced
umbracoumbraco
2

mcp-patterns

Provide patterns for building MCP tools with the Umbraco MCP server SDK.

Official
Intermediate
MCPJamMCPJam
2.1k

create-mcp-eval

Generate Jest or Vitest evaluation tests for MCP servers using @mcpjam/sdk.

Official
Advanced