Skip to main content

Workflow resolver nodes with AI

Leverage AI to identify and execute relevant workflows.

In Actioner, we've introduced a powerful feature called Workflow resolver nodes that takes your workflows to the next level with AI. This feature allows you to identify and execute the most relevant workflows based on the given content, streamlining your automation processes and improving user interactions.

What are workflow resolver nodes?​

Workflow resolver nodes are intelligent components within your workflows that use artificial intelligence to identify the most suitable workflow based on the provided content. These nodes are designed to enhance the automation of your processes, ensuring that the right workflows are triggered at the right time.

Key parameters in workflow resolver nodes​

Workflow resolver nodes accept several essential parameters to customize their functionality:

  1. Content: A plain text field typically representing the user's message or input, often originating from platforms like Slack.

  2. App ID: The unique identifier of the app to where the resolved workflow is added. You can use {{app.id}} to provide the ID of the app that you'll resolve the workflow from. To get the ID of another app, you can use Search apps action from Actioner API integration.

  3. Tag filter: A field that enables you to filter and select workflows based on their tags, ensuring precise targeting.

  4. Messages: This parameter allows you to specify additional context or messages related to the target workflow's usage. Its exact functionality may vary depending on your specific use case.

  5. Additional context: You can provide additional context in JSON format, further enhancing the targeted workflow's ability to execute efficiently. For instance, you can supply inputs that the resolved workflow will require.

How workflow resolver nodes work​

Here's a simplified breakdown of how workflow resolver nodes operate:

  1. Content analysis: The workflow resolver nodes analyzes the provided content, often a user's message, using AI algorithms to understand its intent and context.

  2. Workflow identification: Based on the analysis, the workflow resolver node identifies the most relevant workflow that should be executed.

  3. Parameter values: The node generates parameter values for the selected workflow, ensuring that it has all the necessary information to execute successfully.

  4. Execution: The matching workflows are identified, and with the supplied parameter values, you can start the execution of the workflow or let your users manually run it through clicking a butotn in Slack.

Workflow resolver nodes response​

When a workflow resolver node operates successfully, it returns a response in the following format:

{
  "workflowId": 12345,
  "workflowName": "Sample workflow name",
  "matchScore": 0.4,
  "appId": 12345,
  "allMandatoryParametersAreSatisfied": true,
  "hasMissingParameterValues": false,
  "parameterValues": "{ name: Joe, surname: Doe }",
  "parameterNames": [
    "name",
    "surname"
  ],
  "otherMatchingWorkflows": [
    {
      "appId": "a1",
      "workflowId": "wf2",
      "matchScore": 76.1
    }
  ]
}

Here's what each parameter in the sample response means:

  • workflowName: Returns the name of the identified workflow based on the given content.
  • workflowId: Returns the unique ID of the workflow.
  • matchScore: Indicates the match score of the Workflow Resolver Node.
  • appId: Returns the ID of the app to which the workflow belongs.
  • parameterNames: Lists the parameter names in the workflow.
  • parameterValues: Provides the values generated for the parameters.
  • hasMissingParameterValues: Indicates whether the response contains any missing parameter values (true or false).
  • allMandatoryParametersAreSatisfied: Indicates whether the response includes values for all mandatory parameters (true or false).

How to configure a workflow resolver node​

To add a new workflow resolver node into your workflow:

  1. Click + button in your canvas and select AI option and then select Workflow resolver.

  2. On the opening left panel, enter the Content. You can use data in your workflow such as data from your inputs, event or from the responses of earlier nodes.

  3. Enter App ID. You can use {{app.id}} to provide the ID of the app that you'll resolve the workflow from. To get the ID of another app, you can use Search apps action from Actioner API integration.

  4. Enter the tags of the targeted workflows on Tag filter field.

  5. Optionally enter value in Messages field. This parameter allows you to specify additional context or messages related to the target workflow's usage. Its exact functionality may vary depending on your specific use case.

  6. You can provide Additional context in JSON format, to further enhance the targeted workflow's ability to execute efficiently. For e.g., you can supply inputs that the resolved workflow will require.