Generate a custom connector with the AI Xtensions Generator

Note: This feature is in beta.

Generate a custom connector with AI to save time and effort when creating a custom connector.

Tell the generator the name of the service to use and the actions you want the connector to perform, and it generates a draft custom connector for you. You can customize and import the custom connector directly into your Nintex Workflow tenant.

Beta Feature. The AI features in Nintex Workflow are beta features that provide new generative-AI experiments. Use is voluntary and for experimental purposes only, and your admin can turn off access to it at any time.

Integration with Azure OpenAI. AI features are enabled by integration with Azure OpenAI Service, which comes with Microsoft's built-in Responsible AI and enterprise-grade Azure security. You can learn more about Azure OpenAI Service here and about Microsoft’s applicable data, privacy, and security measures here. You acknowledge and agree that Nintex leverages Azure OpenAI Service to provide these AI features. By using AI features, you authorize and agree that Azure OpenAI Service may access, use, and store any Inputs and Outputs pursuant to its relevant terms and privacy policies.

Modification of Access and Use. Nintex may limit, modify, suspend, discontinue, or terminate your access to AI features at any time, with or without cause (without notice to you). Nintex may also separate these AI features from the Services at any time (without notice to you). Or Nintex may require payment for future or continued use (with notice to you). Nintex is not in any way liable to you for any limitation, modification, suspension, discontinuation, termination, or separation. Without limitation, the laws and regulations governing artificial intelligence and related technology are uncertain and evolving, and the ability to use these AI features may be adversely impacted in the future.

Inputs and Sensitive Information. You are responsible for ensuring that any content you make available while using the AI features (collectively, “Inputs”) is appropriate and permissible under Nintex’s Customer Use Policy and that you have the right to use such Inputs. In using the AI features, you are advised to safeguard important data; to not input confidential, personal, or sensitive information; and to use caution, particularly in relation to the Inputs.

Outputs and Verification. You are also responsible for any information, materials, or other content you generate using AI features (collectively, “Outputs”). You acknowledge that Outputs may not be accurate, complete, reliable, or otherwise fit for use (including from a legal and/or business perspective). Any Outputs should only be used as a starting point, and you should independently verify and cross-reference the Outputs with other reliable sources to ensure that the information or content is accurate, lawful, and otherwise appropriate and permissible under Nintex’s Customer Use Policy and before making any decisions or taking any actions based on the Outputs. You understand and agree that Outputs may not be unique across users and the AI features could generate the same or similar Output for other users. In addition, other users may input similar prompts and receive the same, similar, or different responses. You are not required to use any Outputs and should modify it as appropriate to ensure compliance with applicable laws, this notice, any applicable agreement with Nintex, and Nintex’s Customer Use Policy. Your use of the AI features and any content resulting from such use is entirely at your own risk. Nintex is not responsible or liable to you for the use of AI features or any Outputs. It is provided as-is without any warranties or guarantees of accuracy.

Acknowledgment. By using AI features, you acknowledge and accept that: (i) AI features integrates with third-party software tools provided by Azure OpenAI Service; (ii) your use is subject to this Disclaimer Notice and the terms of any applicable legal agreement between you and Nintex; and (iii) you have read, understood, and agree to this Disclaimer Notice.

Please contact us if you have any questions or concerns regarding this Disclaimer Notice or use of AI features.

Before you start 

Generating a custom connector saves you from typing the full OpenAPI Specification, but you'll still need to review the draft against the third-party service's API. Before you generate your custom connector:

  1. Make sure you're familiar with creating a custom connector using the OpenAPI Specification.
    You must review the OpenAPI Specification JSON before you publish your custom connector.
    See Build your first custom connector for a refresher on building custom connectors.
  2. Check the API documentation for the third-party service to ensure that it is compatible with Nintex Workflow.
    See Limitations and Known Issues.
  3. Consider the actions you want the custom connector to include and which operations you expect to use.
    Keep these notes and the API documentation handy. They'll help you review the generated custom connector.

To generate a custom connector:

  1. Open your Nintex Workflow tenancy.
  2. Click Xtensions in the left navigation.
  3. Click   in the Custom connector list.
  4. Click Start with AI.
  5. Type a description of the custom connector you want to generate, including the name of the service and the actions it should perform.
    For example: From xe.com, convert one currency into another.
  6. When the list of generated operations appears, clear the checkbox for any operations you don't want to include.
    You can customize or hide the action in the next step.
  7. Click Next.
  8. Review the generated OpenAPI Specification carefully against the API documentation:
    • Check that the host and basePath are correct.
    • Check that each path is defined with the correct HTTP method, parameters, and response.
    • Check that the data type is correct for each parameter and response payload, and that all mandatory parameters are defined as required.
      Some parameters or responses may need specification extensions such as x-ntx-dynamic-values and x-ntx-dynamic-schema to handle dynamic data. You may need to add supporting operations to retrieve data.
    • Check that the security definition is correct.
      If you're using OAuth authentication, make sure your scopes are complete, including the offline_access scope to request refresh tokens.
    • Check that any generated text in descriptions and summaries is correct.
    • Make sure you don't modify or remove the three specification extension fields at the bottom of the file:
      • x-ntx-contract-id
      • x-ntx-render-version
      • x-ntx-generated-by-ai
  9. Optionally, customize the actions:
  10. Click Next.
  11. Review and complete the security configuration for the custom connector.
  12. Click Next.
  13. Either select an icon or upload an image to use for the custom connector.
    The custom connector name and description are automatically populated from the generated JSON. You can edit these if you want to.
  14. Click Publish.