tools

Create custom MCP tool handlers with Ruby metadata and call methods.

Updated Mar 13, 2026
One-click install
npx skills add https://github.com/dickdavis/model-context-protocol-rb-plugin --skill tools-dickdavis
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tools
Source: https://github.com/dickdavis/model-context-protocol-rb-plugin/tree/main/skills/tools
Command: npx skills add https://github.com/dickdavis/model-context-protocol-rb-plugin --skill tools-dickdavis

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides the foundational knowledge and examples for creating custom tool handlers within the Model Context Protocol (MCP) framework, enabling developers to build sophisticated AI-powered applications.

Core Features & Use Cases

  • Tool Definition: Learn how to define tool metadata, including name, title, description, and input/output schemas.
  • Response Handling: Implement various response types like text, structured content, images, audio, and embedded resources.
  • Advanced Features: Understand how to incorporate annotations, security schemes, cancellation, progress notifications, and robust error handling.
  • Use Case: Develop a tool that fetches real-time stock prices, a calculator that performs complex mathematical operations, or a content generation tool that drafts email responses.

Quick Start

Use the tools skill to create a tool that fetches weather data for a given location.

Frequently Asked Questions about tools

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

FAQPage Schema
How do I build custom MCP tool handlers in Ruby?

To build MCP tool handlers in Ruby, you define tool metadata including name, title, description, and input/output schemas, then implement call methods to execute actions and return structured or unstructured data.

What response types can an MCP tool handler return?

MCP tool handlers can return diverse response types including text, structured content, images, audio, and embedded resources, enabling AI agents to output both structured and unstructured data formats.

Does the Model Context Protocol support error handling and progress notifications in Ruby tools?

Yes, MCP tool handlers in Ruby support advanced features like annotations, security schemes, cancellation, progress notifications, and robust error handling for sophisticated AI agent integration.

What's the best way to define input and output schemas for an MCP tool?

The best way to define MCP tool schemas is by specifying input and output schemas directly within the tool metadata structure, which validates data passed to your Ruby tool handler methods.

Can I use MCP tool handlers to create an AI agent that fetches real-time data?

Yes, you can develop MCP tool handlers to build AI agents capable of performing specific actions like fetching real-time stock prices, calculating complex math operations, or generating draft email responses.