❗Attention

As of February 2024, the Azure Certified Device program has been retired.

Therefore, Microsoft is no longer accepting submissions of DTDL models to the iot-plugandplay-models repository.

You can still search for and retrieve existing DTDL models from https://devicemodels.azure.com and use DMR ecosystem tools such as Microsoft.IoT.ModelsRepository.CommandLine (aka dmr-client) or the ModelsRepositoryClient SDK.

---

Azure/iot-plugandplay-models repository

This repository includes DTDL models that are made publicly available on https://devicemodels.azure.com. These models can be used to create Azure IoT Plug and Play solutions.

Related tools, samples, and specs can be found in the Azure/iot-plugandplay-models-tools repo. The current repo only stores DTDL models.

Install the dmr-client command line tool

The tool used to validate the models during the PR checks can also be used to add and validate the DTDL interfaces locally.

The Device Models Repository command line tool (aka dmr-client) is published on NuGet and requires dotnet sdk 3.1.x, 5.0.x or 6.0.x.

You can use the dotnet command line via the dotnet tool install command to install dmr-client. The following is an example to install dmr-client as a global tool:

dotnet tool install -g Microsoft.IoT.ModelsRepository.CommandLine --version 1.0.0-beta.6

To learn how to install dmr-client in a local context, please see this guide.

Note: Previous versions of the tool (prior to 1.0.0-beta.3) must first be uninstalled with dotnet tool uninstall -g dmr-client. Use the dotnet tool list -g command to check the id of the tool you want to uninstall.

Update

To update the dmr-client tool (assuming a global install) you can run the following command:

dotnet tool update -g Microsoft.IoT.ModelsRepository.CommandLine --version [target version]

Uninstall

To uninstall the dmr-client tool (assuming a global install) you can run the following command:

dotnet tool uninstall -g Microsoft.IoT.ModelsRepository.CommandLine

Usage of dmr-client

After installing Microsoft.IoT.ModelsRepository.CommandLine the following models repository management commands should be available for usage via the dmr-client alias.

dmr-client
  Microsoft IoT Models Repository CommandLine v1.0.0-beta.6

Usage:
  dmr-client [options] [command]

Options:
  --debug         Shows additional logs for debugging. [default: False]
  --silent        Silences command output on standard out. [default: False]
  --version       Show version information
  -?, -h, --help  Show help and usage information

Commands:
  export    Exports a model producing the model and its dependency chain in an expanded format. 
            The target repository is used for model resolution.
  validate  Validates the DTDL model contained in a file. When validating a single model object the target repository
            is used for model resolution. When validating an array of models only the array contents is used for resolution.
  import    Imports models from a model file into the local repository. The local repository is used for model resolution.
            Target model files for import will first be validated to ensure adherence to IoT Models Repository conventions.
  index     Builds a model index file from the state of a target local models repository.
  expand    For each model in a local repository, generate expanded model files and insert them in-place.
            The expanded version of a model includes the model with its full model dependency chain.

Command Guide

Import a Model to the dtmi/ folder

If you have your model already stored in json files, you can use the dmr-client import command to add those to the dtmi/ folder with the right file name.

Run this from the local repo root folder

dmr-client import --model-file "MyThermostat.json"

This command will rename and locate the file in the appropriate folder

Validate Models

You can validate your models with the dmr-client validate command.

To validate a model file using the DTDL parser.

dmr-client validate --model-file ./my/model/file.json

Note: The validation uses the latest DTDL parser version to ensure all the interfaces are compatible with the DTDL language spec

To validate external dependencies, those must exist in the local repo. To validate those you can specify a local or remote folder to validate against.

Validating a model file using the DTDL parser checking dependencies with the current folder as a local repo.

dmr-client validate --model-file ./my/model/file.json --repo .

Strict validation

The Device Model Repo includes additional requirements, these can be validated with the strict flag.

Validating a model file using the DTDL parser checking dependencies with the current folder as a local repo in strict mode.

dmr-client validate --model-file ./my/model/file.json --repo . --strict

Export models

Models can be exported from a given repo (local or remote) to a single file using a JSON Array.

Retrieving an interface from a custom repo by DTMI:

dmr-client export --dtmi "dtmi:com:example:Thermostat;1" --repo https://raw.githubusercontent.com/Azure/iot-plugandplay-models/main

Create Index

The model repo can host a index.json file with all the ids avaialble in the repository. Read the Index Spec

Building a model index for the repository. If models exceed the page limit new page files will be created relative to the root index.

dmr-client index --local-repo .

Building a model index with a custom page limit indicating max models per page.

dmr-client index --local-repo . --page-limit 100

Create Expanded files

To expand all models from the root directory of a local models repository following Azure IoT conventions. Expanded models are inserted in-place.

dmr-client expand --local-repo .

The default --local-repo value is the current directory. Be sure to specify the root for --local-repo.

dmr-client expand

IoT Models Repository SDKs

Azure SDKs focused on models repository consumption are available in the following languages:

Platform	Package	Source	Samples
.NET	Azure.IoT.ModelsRepository	Source	Samples
Java	com.azure/azure-iot-modelsrepository	Source	Samples
Node	@azure/iot-modelsrepository	Source	Samples
Python	azure-iot-modelsrepository	Source	Samples

Fetch models without any SDK

Any HTTP client can consume the models by applying the repository convention to convert a DTMI to a relative path:

Eg, the interface:

dtmi:azure:DeviceManagement:DeviceInformation;1

can be retrieved from here:

https://devicemodels.azure.com/dtmi/azure/devicemanagement/deviceinformation-1.json

There are samples for .NET and Node in the Azure/iot-plugandplay-models-tools GitHub repository with code you can use to acquire models for your custom IoT solution.