Enable AI Features

This guide walks you through enabling AI features in Nx Cloud for both cloud-hosted and on-prem installations.

Nx Cloud Hosted Setup

To enable AI features for your organization, go to your organization's settings on Nx Cloud and select the organization where you want to enable AI.

In the settings menu, find the "AI Features" section and toggle it to "On".

Ensure that you accept the AI terms to start using the AI features.

AI Features Availability

AI features are available on Hobby, Team and Enterprise Nx Cloud plans.

Self-Healing CI for Enterprise On-Prem Installations

To enable Self-Healing CI for enterprise on-prem installations, follow these steps:

Setup

Add the following configuration to your helm-values.yaml file:

nxApi
  deployment:
    env:
      - name: NX_CLOUD_AI_ENABLED
        value: 'true'
      - name: NX_CLOUD_AI_TOKEN_PROVIDER_TYPE
        value: 'fixed'
      - name: NX_CLOUD_ANTHROPIC_API_KEY
        value: 'sk-ant-...'

frontend
  deployment:
    env:
      - name: NX_CLOUD_AI_ENABLED
        value: 'true'

Then, enable AI features in the organization settings and enable Self-Healing CI in the workspace settings.

Running Self-Healing CI On-Prem

Automatic DTE Agents

If using DTE agents, the self-healing CI step will be automatically added when the setting is enabled in the workspace settings.

Manual DTE

For manual DTE configurations, the nx fix-ci command must be included in the agent configuration after running nx tasks.

Important: This command must run always, meaning that even when previous nx tasks fail, the fix-ci command should still execute.

Non-DTE

For non-DTE setups, refer to the Self-Healing CI documentation for detailed configuration instructions.

Explain with AI for Enterprise On-Prem Installations

To enable the Explain with AI feature for enterprise on-prem installations, add the following configuration to your helm-values.yaml file:

nxApi
  deployment:
    env:
      - name: NX_CLOUD_AI_ENABLED
        value: 'true'

frontend
  deployment:
    env:
      - name: OPENAI_SECRET_KEY
        valueFrom:
          secretKeyRef:
            name: open-ai-secrets
            key: OPENAI_SECRET_KEY
      - name: NX_CLOUD_AI_ENABLED
        value: 'true'

(We use gpt-4o for the Explain with AI feature)

Then, enable AI features in the organization settings.

Regional Availability

AI features are not available for the EU cluster in public cloud installations due to regional restrictions. However, on-prem customers in the EU can still use these features by providing their own OpenAI secret key (for Explain with AI) or Anthropic API key (for Self-Healing CI) and enabling the required environment variables.