Schema Reference

JSON schemas for AI-assisted configuration

Overview

intu provides JSON schemas for its YAML configuration files. These schemas enable IDE autocompletion, real-time validation, and allow AI tools to generate correct configuration without guesswork.

Two schemas are available:

• Channel Schema — validates channel.yaml files inside each channel directory
• Profile Schema — validates the root intu.yaml and profile override files

Channel Schema

The channel schema defines the structure of channel.yaml files, including source configuration, transformer and validator references, destination mappings, tags, groups, and all other channel-level properties.

VS Code setup

Add the following to your VS Code settings.json (workspace or user level) to enable autocompletion and validation for channel files. Requires the YAML extension.

{
  "yaml.schemas": {
    "https://intu.dev/schema/channel.schema.json": "channels/*/channel.yaml"
  }
}

What the schema covers

• Channel metadata: id, name, description, enabled, tags, group
• Source configuration for all supported source types
• Transformer and validator file references
• Destination references (by name) and inline destination config
• Retry, error handling, and logging options

Profile Schema

The profile schema defines the structure of the root intu.yaml file and any profile override files (intu.dev.yaml, intu.prod.yaml, etc.). It covers runtime settings, global Kafka configuration, named destinations, and all top-level properties.

VS Code setup

{
  "yaml.schemas": {
    "https://intu.dev/schema/profile.schema.json": ["intu.yaml", "intu.*.yaml"]
  }
}

Combined VS Code configuration

To enable both schemas at once, combine them in a single settings.json block:

{
  "yaml.schemas": {
    "https://intu.dev/schema/channel.schema.json": "channels/*/channel.yaml",
    "https://intu.dev/schema/profile.schema.json": ["intu.yaml", "intu.*.yaml"]
  }
}

Using with AI

The JSON schemas are designed to be consumed by AI assistants and code generation tools. When provided as context, they give the AI a complete specification of every valid property, its type, allowed values, and description.

This means an AI assistant can:

• Generate valid channel.yaml files from natural language descriptions
• Suggest correct configuration values with proper types and constraints
• Validate existing configuration and identify errors
• Explain what each configuration option does

Providing schema context

When working with an AI assistant, include the schema URL or content as part of your prompt. Most AI-powered coding tools can fetch and parse JSON schemas automatically.

Using the intu channel schema at https://intu.dev/schema/channel.schema.json,
generate a channel.yaml for an HL7 ADT listener on port 2575 that transforms
messages and sends them to a Kafka topic.