| layout | default |
|---|---|
| title | Chapter 3: Tools, Prompts, Resources, and Schema Discipline |
| nav_order | 3 |
| parent | MCP Ruby SDK Tutorial |
Welcome to Chapter 3: Tools, Prompts, Resources, and Schema Discipline. In this part of MCP Ruby SDK Tutorial: Building MCP Servers and Clients in Ruby, you will build an intuitive mental model first, then move into concrete implementation details and practical production tradeoffs.
This chapter focuses on modeling MCP primitives with predictable behavior and schema quality.
- design tool schemas that validate arguments and output shapes reliably
- structure prompt and resource handlers for maintainable growth
- use annotations and metadata consistently across primitives
- avoid schema drift that breaks client interoperability
- define explicit tool argument schemas and output contracts
- model prompt arguments with stable names and titles
- expose resource identifiers and templates with clear naming patterns
- test edge cases for empty, invalid, and over-broad input payloads
- Ruby SDK README - Tools
- Ruby SDK README - Tool Output Schemas
- Ruby SDK README - Prompts
- Ruby SDK README - Resources
You now have a schema-first primitive strategy for Ruby MCP servers.
Next: Chapter 4: Notifications, Logging, and Observability
The dev module in dev.yml handles a key part of this chapter's functionality:
name: mcp-ruby
type: ruby
up:
- ruby
- bundler
commands:
console:
desc: Open console with the gem loaded
run: bin/console
build:
desc: Build the gem using rake build
run: bin/rake build
test:
desc: Run tests
syntax:
argument: file
optional: args...
run: |
if [[ $# -eq 0 ]]; then
bin/rake test
else
bin/rake -I test "$@"
fi
style:
desc: Run rubocop
aliases: [rubocop, lint]
run: bin/rake rubocop
This module is important because it defines how MCP Ruby SDK Tutorial: Building MCP Servers and Clients in Ruby implements the patterns covered in this chapter.
The streamable_http_server module in examples/streamable_http_server.rb handles a key part of this chapter's functionality:
# frozen_string_literal: true
$LOAD_PATH.unshift(File.expand_path("../lib", __dir__))
require "mcp"
require "rack/cors"
require "rackup"
require "json"
require "logger"
# Create a logger for SSE-specific logging
sse_logger = Logger.new($stdout)
sse_logger.formatter = proc do |severity, datetime, _progname, msg|
"[SSE] #{severity} #{datetime.strftime("%H:%M:%S.%L")} - #{msg}\n"
end
# Tool that returns a response that will be sent via SSE if a stream is active
class NotificationTool < MCP::Tool
tool_name "notification_tool"
description "Returns a notification message that will be sent via SSE if stream is active"
input_schema(
properties: {
message: { type: "string", description: "Message to send via SSE" },
delay: { type: "number", description: "Delay in seconds before returning (optional)" },
},
required: ["message"],
)
class << self
attr_accessor :logger
def call(message:, delay: 0)
sleep(delay) if delay > 0
logger&.info("Returning notification message: #{message}")This module is important because it defines how MCP Ruby SDK Tutorial: Building MCP Servers and Clients in Ruby implements the patterns covered in this chapter.
The .rubocop module in .rubocop.yml handles a key part of this chapter's functionality:
inherit_gem:
rubocop-shopify: rubocop.yml
plugins:
- rubocop-minitest
- rubocop-rake
AllCops:
TargetRubyVersion: 2.7
Gemspec/DevelopmentDependencies:
Enabled: true
Lint/IncompatibleIoSelectWithFiberScheduler:
Enabled: true
Minitest/LiteralAsActualArgument:
Enabled: true
This module is important because it defines how MCP Ruby SDK Tutorial: Building MCP Servers and Clients in Ruby implements the patterns covered in this chapter.
flowchart TD
A[dev]
B[streamable_http_server]
C[.rubocop]
A --> B
B --> C