Creating Completely New Custom Rulesets

To create a completely new custom ruleset that isn’t based on an existing one, use one of the following approaches:

Search the MuleSoft Ideas Portal for ideas for new rulesets. If you don’t find your idea there, consider submitting your idea in the MuleSoft Ideas Portal.
Generate a ruleset using Anypoint CLI commands by scaffolding it from a specification’s data schema. See Generating Rulesets from a Specification.
Create a ruleset using the ruleset development CLI commands, see Create Rulesets Using the Ruleset Development CLI.
Create a ruleset using the ruleset modeling language, see Create Rulesets Using the Modeling Language.

After you create the ruleset, see Validating and Publishing Custom Rulesets.

As with custom code and configurations, rulesets are not considered supported MuleSoft products. For help with issues with custom rulesets, post an issue in the AMF Custom Validator Github repository.

Generating Rulesets from a Specification

You can generate a starter ruleset from specifications' data schemas using the following command sequence:

Inspect API Specifications for API Type
Initialize Rulesets from Data Schemas

Inspect API Specifications for API Type

Use the following command to inspect API specifications for API type. This helps you determine which data schema to use.

governance:api:inspect

> governance:api:inspect [flags] <api-specification>

Inspects the API specification passed in api-specification and lists all its schemas, such as headers, requests, and response payloads. You can use this schema information in the governance:ruleset:init command. See governance:ruleset:init.

This command accepts the default flags.

Example command:

anypoint-cli-v4 governance:api:inspect my-healthcare-api.yaml

Example schema

types:
  patientmultipleBirthBoolean:
    properties:
      multipleBirthBoolean:
        description: Whether patient is part of a multiple birth
        type: boolean
  patientmultipleBirthInteger:
    properties:
      multipleBirthInteger:
        description: Whether patient is part of a multiple birth
        type: integer

        .
        .
        .

  PatientEntry:
    type: FHIR_commons.Entry
    properties:
      resource: Patient

  PatientBundle:
    type: FHIR_commons.Bundle
    properties:
      entry?: PatientEntry[]

Example output:

  'patientmultipleBirthBoolean',
  'PatientBundle',
  'patientmultipleBirthInteger',
  'PatientEntry'

Initialize Rulesets from Data Schemas

Use the following command to initialize rulesets from data schemas.

governance:ruleset:init

> governance:ruleset:init [flags] <schema>

Initializes a ruleset based on the data schema passed in the schema parameter.

In addition to the default flags, this command accepts the following flags:

Flag Description

Flag	Description
`--types <types>`	The `types` flag gives the target types to export as rules. You can use the `governance:api:inspect` command to identify the types to specify for this flag. See governance:api:inspect.
`--name <name>`	The `name` flag is the name of the ruleset. This defaults to `GeneratedRuleset`.

--types <types>

The types flag gives the target types to export as rules. You can use the governance:api:inspect command to identify the types to specify for this flag. See governance:api:inspect.

--name <name>

The name flag is the name of the ruleset. This defaults to GeneratedRuleset.

Example command:

anypoint-cli-v4 governance:ruleset:init --types patientmultipleBirthBoolean,patientBundle,patientmultipleBirthInteger --name=my-ruleset mydataschema

Create Rulesets Using the Ruleset Development CLI

An open-source version of ruleset development CLI is available on npm. It is provided and documented separately from the Anypoint CLI governance commands.

To create rulesets using the ruleset development CLI:

Install ruleset-development-cli from npmjs.com.
Follow the AMF Rulesets tutorial to create a ruleset.

Create Rulesets Using the Modeling Language

To create rulesets using the modeling language, read the AMF Custom Validation section in the AML Open Source project to learn how to write a ruleset using the modeling language.