Posit Assistant

Workbench | Preview

WarningPreview feature

This feature is in preview. Preview features are unsupported and may face breaking changes in a future release. Any issues found in the feature will be addressed during the regular release schedule; they will not result in immediate patches or hotfixes.

We encourage customers to try these features and we welcome any feedback via Posit Support, but we recommend that the feature not be used in production until it is in general availability (i.e., officially released as a full feature). To provide feedback, please email your Posit Customer Success representative or and specify that you are trialing this feature.

Posit Assistant is an AI coding assistant built specifically for data science workflows. Users can configure model providers like Anthropic or GitHub Copilot, or administrators can configure Workbench-managed credentials for Amazon Bedrock or Snowflake Cortex.

This page covers configuring Posit Assistant for Positron Pro sessions. For end-user documentation, see the Posit Assistant documentation.

Enable or disable Posit Assistant

Posit Assistant is enabled by default. Control its availability with the assistant-enabled profile key in /etc/rstudio/profiles, which enforces Posit Assistant across both Positron Pro and RStudio Pro sessions. See Posit Assistant Admin Controls for details.

Language model providers

Posit Assistant supports multiple language model providers, including Anthropic, GitHub Copilot, Amazon Bedrock, Snowflake Cortex, and Microsoft Foundry. Enable or disable providers and constrain their available models through the ai-providers-enforced profile setting. See AI Provider Admin Controls for the full list of recognized providers and configuration options.

Some providers require additional Workbench setup before they can be used:

Amazon Bedrock

To set up Amazon Bedrock with Posit Assistant:

  1. Configure AWS Credentials for Workbench, if not already configured.
  2. Configure AWS Identity and access management (IAM) for Amazon Bedrock policies to allow access to Bedrock models.
  3. Enable the bedrock provider and constrain its available models through the ai-providers-enforced profile setting. See AI Provider Admin Controls, which also covers the provider’s aws connection settings.

Snowflake Cortex

To set up Snowflake Cortex with Posit Assistant:

  1. Integrate Workbench with Snowflake.
  2. Configure Snowflake access to Snowflake Cortex.
  3. Enable the snowflake-cortex provider and constrain its available models through the ai-providers-enforced profile setting. See AI Provider Admin Controls, which also covers the provider’s snowflake connection settings.

Microsoft Foundry

Microsoft Foundry provides access to AI models, including Azure OpenAI models. Posit Assistant uses Workbench delegated Azure credentials with Entra ID to authenticate. Unlike other providers, Microsoft Foundry requires the following Workbench extension settings in addition to any configuration you supply through AI Provider Admin Controls.

To set up Microsoft Foundry with Posit Assistant:

  1. Configure delegated Azure credentials for Workbench, if not already configured.

  2. Set the Microsoft Foundry endpoint. This is the Azure resource endpoint URL (for example, https://my-resource.openai.azure.com or https://my-resource.services.ai.azure.com):

    /etc/rstudio/positron-user-settings.json
    {
  "posit.workbench.foundry.endpoint": "https://my-resource.services.ai.azure.com"
}

    This can also be set using Positron Enforced Settings:

    /etc/rstudio/enforced-settings.json
    {
  "posit.workbench.foundry.endpoint": "https://my-resource.services.ai.azure.com"
}

  3. If a gateway fronts Foundry with its own Entra ID app registration, set posit.workbench.foundry.audience to the gateway app’s identifier. The default is https://cognitiveservices.azure.com/ for direct-to-Foundry deployments:

    /etc/rstudio/positron-user-settings.json
    {
  "posit.workbench.foundry.endpoint": "https://my-gateway.example.com",
  "posit.workbench.foundry.audience": "api://example-gateway-app-id"
}

  4. Enable the ms-foundry provider through the ai-providers-enforced profile setting. See AI Provider Admin Controls.

By default, Microsoft Foundry uses model-router for server-side model selection. If you are not using model-router, turn off model discovery and declare the available models with the provider’s custom model list in the file that ai-providers-enforced references:

/etc/rstudio/ai-providers-enforced/default.json
{
  "providers": {
    "ms-foundry": {
      "enabled": true,
      "models": {
        "discovery": "off",
        "custom": [
          {
            "id": "gpt-4o",
            "name": "GPT-4o",
            "maxContextLength": 128000,
            "supportsTools": true,
            "supportsImages": true,
            "supportsToolResultImages": false,
            "supportsWebSearch": false
          }
        ]
      }
    }
  }
}

Other providers

Refer to the Posit Assistant provider documentation for information on configuring additional language model providers.

Back to top