Azure Functions Core Tools lets you develop and test your functions on your local computer. When you're ready, you can also use Core Tools to deploy your code project to Azure and work with application settings.

::: zone pivot="programming-language-csharp"

You're viewing the C# version of this article. Make sure to select your preferred Functions programming language at the top of the article.

If you want to get started right away, complete the Core Tools quickstart article. ::: zone-end ::: zone pivot="programming-language-java"

You're viewing the Java version of this article. Make sure to select your preferred Functions programming language at the top of the article.

If you want to get started right away, complete the Core Tools quickstart article. ::: zone-end ::: zone pivot="programming-language-javascript"

You're viewing the JavaScript version of this article. Make sure to select your preferred Functions programming language at the top of the article.

If you want to get started right away, complete the Core Tools quickstart article. ::: zone-end ::: zone pivot="programming-language-powershell"

You're viewing the PowerShell version of this article. Make sure to select your preferred Functions programming language at the top of the article.

If you want to get started right away, complete the Core Tools quickstart article. ::: zone-end ::: zone pivot="programming-language-python"

You're viewing the Python version of this article. Make sure to select your preferred Functions programming language at the top of the article.

If you want to get started right away, complete the Core Tools quickstart article. ::: zone-end ::: zone pivot="programming-language-typescript"

You're viewing the TypeScript version of this article. Make sure to select your preferred Functions programming language at the top of the article.

If you want to get started right away, complete the Core Tools quickstart article. ::: zone-end

[!INCLUDE functions-install-core-tools]

For help with version-related issues, see Core Tools versions.

::: zone pivot="programming-language-python"

::: zone pivot="programming-language-csharp"

func init MyProjFolder --worker-runtime dotnet-isolated 

By default this command creates a project that runs in-process with the Functions host on the current Long-Term Support (LTS) version of .NET Core. You can use the --target-framework option to target a specific supported version of .NET, including .NET Framework. For more information, see the func init reference.

func init MyProjFolder --worker-runtime dotnet 

This command creates a project that runs on the current Long-Term Support (LTS) version of .NET Core. For other .NET version, create an app that runs in an isolated worker process from the Functions host.

For a comparison between the two .NET process models, see the process mode comparison article. ::: zone-end ::: zone pivot="programming-language-java" Java uses a Maven archetype to create the local project, along with your first HTTP triggered function. Rather than using func init and func new, you should instead follow the steps in the Command line quickstart.
::: zone-end ::: zone pivot="programming-language-javascript"

func init MyProjFolder --worker-runtime javascript --model V4

func init MyProjFolder --worker-runtime javascript --model V3

This command creates a JavaScript project that uses the desired programming model version. ::: zone-end
::: zone pivot="programming-language-typescript"

func init MyProjFolder --worker-runtime typescript --model V4

func init MyProjFolder --worker-runtime typescript --model V3

This command creates a TypeScript project that uses the desired programming model version. ::: zone-end
::: zone pivot="programming-language-powershell"

func init MyProjFolder --worker-runtime powershell

::: zone-end ::: zone pivot="programming-language-python"

func init MyProjFolder --worker-runtime python --model V2

func init MyProjFolder --worker-runtime python

This command creates a Python project that uses the desired programming model version. ::: zone-end

When you run func init without the --worker-runtime option, you're prompted to choose your project language. To learn more about the available options for the func init command, see the func init reference.

To add a function to your project, run the func new command using the --template option to select your trigger template. The following example creates an HTTP trigger named MyHttpTrigger:

func new --template "Http Trigger" --name MyHttpTrigger

This example creates a Queue Storage trigger named MyQueueTrigger:

func new --template "Azure Queue Storage Trigger" --name MyQueueTrigger

The following considerations apply when adding functions:

• When you run func new without the --template option, you're prompted to choose a template.

• Use the func templates list command to see the complete list of available templates for your language.

• When you add a trigger that connects to a service, you'll also need to add an application setting that references a connection string or a managed identity to the local.settings.json file. Using app settings in this way prevents you from having to embed credentials in your code. For more information, see Work with app settings locally. ::: zone pivot="programming-language-csharp"

• Core Tools also adds a reference to the specific binding extension to your C# project. ::: zone-end

To learn more about the available options for the func new command, see the func new reference.

Functions provides a set of service-specific input and output bindings, which make it easier for your function to connection to other Azure services without having to use the service-specific client SDKs. For more information, see Azure Functions triggers and bindings concepts.

To add an input or output binding to an existing function, you must manually update the function definition. [!INCLUDE functions-add-output-binding-example-all-langs] The following considerations apply when adding bindings to a function: ::: zone pivot="programming-language-javascript,programming-language-typescript,programming-language-python,programming-language-powershell"

• For languages that define functions using the function.json configuration file, Visual Studio Code simplifies the process of adding bindings to an existing function definition. For more information, see Connect functions to Azure services using bindings. ::: zone-end
• When you add bindings that connect to a service, you must also add an application setting that references a connection string or managed identity to the local.settings.json file. For more information, see Work with app settings locally.
  ::: zone pivot="programming-language-java,programming-language-javascript,programming-language-typescript,programming-language-powershell"
• When you add a supported binding, the extension should already be installed when your app uses extension bundle. For more information, see extension bundles. ::: zone-end
  ::: zone pivot="programming-language-csharp"
• When you add a binding that requires a new binding extension, you must also add a reference to that specific binding extension in your C# project. ::: zone-end
  ::: zone pivot="programming-language-csharp" For more information, including links to example binding code that you can refer to, see Add bindings to a function.
  ::: zone-end
  ::: zone pivot="programming-language-java" For more information, including links to example binding code that you can refer to, see Add bindings to a function.
  ::: zone-end
  ::: zone pivot="programming-language-javascript" For more information, including links to example binding code that you can refer to, see Add bindings to a function.
  ::: zone-end
  ::: zone pivot="programming-language-powershell"
  For more information, including links to example binding code that you can refer to, see Add bindings to a function.
  ::: zone-end
  ::: zone pivot="programming-language-python"
  For more information, including links to example binding code that you can refer to, see Add bindings to a function.
  ::: zone-end ::: zone pivot="programming-language-typescript" For more information, including links to example binding code that you can refer to, see Add bindings to a function.
  ::: zone-end

Before you can run or debug the functions in your project, you need to start the Functions host from the root directory of your project. The host enables triggers for all functions in the project. Use this command to start the local runtime:

::: zone pivot="programming-language-java"

mvn clean package 
mvn azure-functions:run

::: zone-end
::: zone pivot="programming-language-csharp,programming-language-javascript,programming-language-powershell,programming-language-python"

func start

::: zone-end
::: zone pivot="programming-language-typescript"

npm install
npm start     

::: zone-end ::: zone pivot="programming-language-python" This command must be run in a virtual environment. ::: zone-end

When the Functions host starts, it outputs a list of functions in the project, including the URLs of any HTTP-triggered functions, like in this example:

Found the following functions:
Host.Functions.MyHttpTrigger

Job host started
Http Function MyHttpTrigger: http://localhost:7071/api/MyHttpTrigger

Keep in mind the following considerations when running your functions locally:

• By default, authorization isn't enforced locally for HTTP endpoints. This means that all local HTTP requests are handled as authLevel = "anonymous". For more information, see Authorization level. You can use the --enableAuth option to require authorization when running locally. For more information, see func start

• You can use the local Azurite emulator when locally running functions that require access to Azure Storage services (Queue Storage, Blob Storage, and Table Storage) without having to connect to these services in Azure. When using local emulation, make sure to start Azurite before starting the local host (func.exe). For more information, see Local storage emulation. ::: zone pivot="programming-language-python"

• You can use local Azurite emulation to meet the storage requirement of the Python v2 worker. ::: zone-end

• You can trigger non-HTTP functions locally without connecting to a live service. For more information, see Run a local function.

• When you include your Application Insights connection information in the local.settings.json file, local log data is written to the specific Application Insights instance. To keep local telemetry data separate from production data, consider using a separate Application Insights instance for development and testing. ::: zone pivot="programming-language-csharp,programming-language-javascript"

• When using version 1.x of the Core Tools, instead use the func host start command to start the local runtime. ::: zone-end

With your local Functions host (func.exe) running, you can now trigger individual functions to run and debug your function code. The way in which you execute an individual function depends on its trigger type.

HTTP triggers are started by sending an HTTP request to the local endpoint and port as displayed in the func.exe output, which has this general format:

http://localhost:<PORT>/api/<FUNCTION_NAME>

In this URL template, <FUNCTION_NAME> is the name of the function or route and <PORT> is the local port on which func.exe is listening.

For example, this cURL command triggers the MyHttpTrigger quickstart function from a GET request with the name parameter passed in the query string:

curl --get http://localhost:7071/api/MyHttpTrigger?name=Azure%20Rocks

This example is the same function called from a POST request passing name in the request body, shown for both Bash shell and Windows command line:

curl --request POST http://localhost:7071/api/MyHttpTrigger --data '{"name":"Azure Rocks"}'

curl --request POST http://localhost:7071/api/MyHttpTrigger --data "{'name':'Azure Rocks'}"

The following considerations apply when calling HTTP endpoints locally:

• You can make GET requests from a browser passing data in the query string. For all other HTTP methods, you must use a HTTP testing tool that also keeps your data secure. For more information, see HTTP test tools.

• Make sure to use the same server name and port that the Functions host is listening on. You see an endpoint like this in the output generated when starting the Function host. You can call this URL using any HTTP method supported by the trigger.

There are two ways to execute non-HTTP triggers locally. First, you can connect to live Azure services, such as Azure Storage and Azure Service Bus. This directly mirrors the behavior of your function when running in Azure. When using live services, make sure to include the required named connection strings in the local settings file. You may consider using a different service connection during development than you do in production by using a different connection string in the local.settings.json file than you use in the function app settings in Azure.

Event Grid triggers require extra configuration to run locally.

You can also run a non-HTTP function locally using REST by calling a special endpoint called an administrator endpoint. Use this format to call the admin endpoint and trigger a specific non-HTTP function:

http://localhost:<PORT>/admin/functions/<FUNCTION_NAME>

In this URL template, <FUNCTION_NAME> is the name of the function or route and <PORT> is the local port on which func.exe is listening.

You can optionally pass test data to the execution in the body of the POST request. To pass test data, you must supply the data in the body of a POST request message, which has this JSON format:

{
    "input": "<TRIGGER_INPUT>"
}

The <TRIGGER_INPUT> value contains data in a format expected by the function. This cURL example is shown for both Bash shell and Windows command line:

curl --request POST -H "Content-Type:application/json" --data '{"input":"sample queue data"}' http://localhost:7071/admin/functions/QueueTrigger

curl --request POST -H "Content-Type:application/json" --data "{'input':'sample queue data'}" http://localhost:7071/admin/functions/QueueTrigger

The previous examples generate a POST request that passes a string sample queue data to a function named QueueTrigger function, which simulates data arriving in the queue and triggering the function

The following considerations apply when using the administrator endpoint for local testing:

• You can call the functions administrator endpoint (http://localhost:{port}/admin/functions/) to return a list of administrator URLs for all available functions, both HTTP triggered and non-HTTP triggered.

• Authentication and authorization are bypassed when running locally. The same APIs exist in Azure, but when you try to call the same administrator endpoints in Azure, you must provide an access key. To learn more, see Work with access keys.

• Access keys are valuable shared secrets. When used locally, they must be securely stored outside of source control. Because authentication and authorization aren't required by Functions when running locally, you should avoid using and storing access keys unless your scenarios require it.

• Calling an administrator endpoint and passing test data is similar to using the Test tab in the Azure portal.

Event Grid triggers have specific requirements to enable local testing. For more information, see Local testing with viewer web app.

The Azure Functions Core Tools supports three types of deployment:

You must have either the Azure CLI or Azure PowerShell installed locally to be able to publish to Azure from Core Tools. By default, Core Tools uses these tools to authenticate with your Azure account.

If you don't have these tools installed, you need to instead get a valid access token to use during deployment. You can present an access token using the --access-token option in the deployment commands.

::: zone pivot="programming-language-csharp,programming-language-javascript,programming-language-powershell,programming-language-python,programming-language-typescript" To publish your local code to a function app in Azure, use the func azure functionapp publish command, as in the following example:

func azure functionapp publish <FunctionAppName>

This command publishes project files from the current directory to the <FunctionAppName> as a .zip deployment package. If the project requires compilation, it's done remotely during deployment. ::: zone-end ::: zone pivot="programming-language-java" Java uses Maven to publish your local project to Azure instead of Core Tools. Use the following Maven command to publish your project to Azure:

mvn azure-functions:deploy

When you run this command, Azure resources are created during the initial deployment based on the settings in your pom.xml file. For more information, see Deploy the function project to Azure. ::: zone-end
::: zone pivot="programming-language-csharp,programming-language-javascript,programming-language-powershell,programming-language-python,programming-language-typescript"
The following considerations apply to this kind of deployment:

• Publishing overwrites existing files in the remote function app deployment.

• You must have already created a function app in your Azure subscription. Core Tools deploys your project code to this function app resource. To learn how to create a function app from the command prompt or terminal window using the Azure CLI or Azure PowerShell, see Create a Function App for serverless execution. You can also create these resources in the Azure portal. You get an error when you try to publish to a <FunctionAppName> that doesn't exist in your subscription.

• A project folder may contain language-specific files and directories that shouldn't be published. Excluded items are listed in a .funcignore file in the root project folder.

• By default, your project is deployed so that it runs from the deployment package. To disable this recommended deployment mode, use the --nozip option.

• A remote build is performed on compiled projects. This can be controlled by using the --no-build option.

• Use the --publish-local-settings option to automatically create app settings in your function app based on values in the local.settings.json file.

• To publish to a specific named slot in your function app, use the --slot option. ::: zone-end

Core Tools lets you deploy your containerized function app to both managed Azure Container Apps environments and Kubernetes clusters that you manage.

Use the following func azurecontainerapps deploy command to deploy an existing container image to a Container Apps environment:

func azurecontainerapps deploy --name <APP_NAME> --environment <ENVIRONMENT_NAME> --storage-account <STORAGE_CONNECTION> --resource-group <RESOURCE_GROUP> --image-name <IMAGE_NAME> [--registry-password] [--registry-server] [--registry-username]

When you deploy to an Azure Container Apps environment, the following considerations apply:

• The environment and storage account must already exist. The storage account connection string you provide is used by the deployed function app.

• You don't need to create a separate function app resource when deploying to Container Apps.

• Storage connection strings and other service credentials are important secrets. Make sure to securely store any script files using func azurecontainerapps deploy and don't store them in any publicly accessible source control systems. You can encrypt the local.settings.json file for added security.

For more information, see Azure Container Apps hosting of Azure Functions.

The following func kubernetes deploy command uses the Dockerfile to generate a container in the specified registry and deploy it to the default Kubernetes cluster.

func kubernetes deploy --name <DEPLOYMENT_NAME> --registry <REGISTRY_USERNAME> 

Azure Functions on Kubernetes using KEDA is an open-source effort that you can use free of cost. Best-effort support is provided by contributors and from the community. To learn more, see Deploying a function app to Kubernetes.

[!INCLUDE functions-local-settings-file]

The following considerations apply when working with the local settings file:

• Because the local.settings.json may contain secrets, such as connection strings, you should never store it in a remote repository. Core Tools helps you encrypt this local settings file for improved security. For more information, see Local settings file. You can also encrypt the local.settings.json file for added security.

• By default, local settings aren't migrated automatically when the project is published to Azure. Use the --publish-local-settings option when you publish your project files to make sure these settings are added to the function app in Azure. Values in the ConnectionStrings section are never published. You can also upload settings from the local.settings.json file at any time.

• You can download and overwrite settings in your local.settings.json file with settings from your function app in Azure. For more information, see Download application settings.
  ::: zone pivot="programming-language-csharp"

• The function app settings values can also be read in your code as environment variables. For more information, see Environment variables. ::: zone-end ::: zone pivot="programming-language-java"

• The function app settings values can also be read in your code as environment variables. For more information, see Environment variables. ::: zone-end ::: zone pivot="programming-language-javascript,programming-language-typescript"

• The function app settings values can also be read in your code as environment variables. For more information, see Environment variables. ::: zone-end ::: zone pivot="programming-language-powershell"

• The function app settings values can also be read in your code as environment variables. For more information, see Environment variables. ::: zone-end ::: zone pivot="programming-language-python"

• The function app settings values can also be read in your code as environment variables. For more information, see Environment variables. ::: zone-end

• When no valid storage connection string is set for AzureWebJobsStorage and a local storage emulator isn't being used, an error is shown. You can use Core Tools to download a specific connection string from any of your Azure Storage accounts.

From the project root, use the following command to download all application settings from the myfunctionapp12345 app in Azure:

func azure functionapp fetch-app-settings myfunctionapp12345

This command overwrites any existing settings in the local.settings.json file with values from Azure. When not already present, new items are added to the collection. For more information, see the func azure functionapp fetch-app-settings command.

Core Tools also make it easy to get the connection string of any storage account to which you have access. From the project root, use the following command to download the connection string from a storage account named mystorage12345.

func azure storage fetch-connection-string mystorage12345

This command adds a setting named mystorage12345_STORAGE to the local.settings.json file, which contains the connection string for the mystorage12345 account. For more information, see the func azure storage fetch-connection-string command.

For improved security during development, consider encrypting the local.settings.json file.

When you publish your project files to Azure without using the --publish-local-settings option, settings in the local.settings.json file aren't set in your function app. You can always rerun the func azure functionapp publish with the --publish-settings-only option to upload just the settings without republishing the project files.

The following example uploads just settings from the Values collection in the local.settings.json file to the function app in Azure named myfunctionapp12345:

func azure functionapp publish myfunctionapp12345 --publish-settings-only

To improve security of connection strings and other valuable data in your local settings, Core Tools lets you encrypt the local.settings.json file. When this file is encrypted, the runtime automatically decrypts the settings when needed the same way it does with application setting in Azure. You can also decrypt a locally encrypted file to work with the settings.

Use the following command to encrypt the local settings file for the project:

func settings encrypt

Use the following command to decrypt an encrypted local setting, so that you can work with it:

func settings decrypt

When the settings file is encrypted and decrypted, the file's IsEncrypted setting also gets updated.

Functions triggers and bindings are implemented as .NET extension (NuGet) packages. To be able to use a specific binding extension, that extension must be installed in the project.

::: zone pivot="programming-language-javascript,programming-language-csharp" This section doesn't apply to version 1.x of the Functions runtime. In version 1.x, supported bindings were included in the core product extension. ::: zone-end

::: zone pivot="programming-language-csharp" For C# class library projects, add references to the specific NuGet packages for the binding extensions required by your functions. C# script (.csx) project must use extension bundles. ::: zone-end ::: zone pivot="programming-language-java,programming-language-javascript,programming-language-powershell,programming-language-python,programming-language-typescript" Functions provides extension bundles to make is easy to work with binding extensions in your project. Extension bundles, which are versioned and defined in the host.json file, install a complete set of compatible binding extension packages for your app. Your host.json should already have extension bundles enabled. If for some reason you need to add or update the extension bundle in the host.json file, see Extension bundles.

If you must use a binding extension or an extension version not in a supported bundle, you need to manually install extensions. For such rare scenarios, see the func extensions install command. ::: zone-end

Major versions of Azure Functions Core Tools are linked to specific major versions of the Azure Functions runtime. For example, version 4.x of Core Tools supports version 4.x of the Functions runtime. This version is the recommended major version of both the Functions runtime and Core Tools. You can determine the latest release version of Core Tools in the Azure Functions Core Tools repository.

Run the following command to determine the version of your current Core Tools installation:

func --version

Unless otherwise noted, the examples in this article are for version 4.x.

The following considerations apply to Core Tools installations:

• You can only install one version of Core Tools on a given computer.

• When upgrading to the latest version of Core Tools, you should use the same method that you used for original installation to perform the upgrade. For example, if you used an MSI on Windows, uninstall the current MSI and install the latest one. Or if you used npm, rerun the npm install command.

• Version 2.x and 3.x of Core Tools were used with versions 2.x and 3.x of the Functions runtime, which have reached their end of support. For more information, see Azure Functions runtime versions overview.
  ::: zone pivot="programming-language-csharp,programming-language-javascript"

• Version 1.x of Core Tools is required when using version 1.x of the Functions Runtime, which is still supported. This version of Core Tools can only be run locally on Windows computers. If you're currently running on version 1.x, you should consider migrating your app to version 4.x today. ::: zone-end

Learn how to develop, test, and publish Azure functions by using Azure Functions core tools. Azure Functions Core Tools is open source and hosted on GitHub. To file a bug or feature request, open a GitHub issue.