Connected System Templates

Introduction

This article provides details of the various types of connected system templates. Each connected system template is custom for the service it connects to, and it will tailor fit integrations that use this connected system with fields specific to that third party system.

Common Properties

Each connected system template has the following properties:

Field Description
System Logo An image document used to visually represent the external system or service being connected to. The logo is displayed in the integration designer and in the process modeler on integrations that connect to this system (using the Call Integration Smart Service). If no document is selected a default logo will be used instead.
Name The name of the connected system. Use a name that will uniquely identify this connection to the external system.
Description Supplemental information about the connected system that is displayed when selecting the system in the integration designer and in the application contents grid.

OpenAPI

Designers can use OpenAPI documents to generate a connected system that can inherit base URLs, authentication details, and a list of operations. Then, using that connected system, the designer can create integrations for those operations. Appian will automatically generate an expression for the URL with rule inputs for any path parameters. It will also generate required headers and query parameters.

In addition to the common properties, an OpenAPI connected system has the following properties:

Field Description
OpenAPI/Swagger Document An OpenAPI documents (formerly known as a Swagger document) defines a standard interface to a RESTful API. They can be read by a human or computer to understand the capabilities of a web service without access to source code/documentation. By uploading these to your connected system, Appian can inherit base URLs, authentication details, and a list of operations.
Base URL A consistent prefix for the URLs of this connected system's integrations. This value supports environment specific configuration for variation across environments. This value can be selected from the list of base URLs that are inherited from the attached OpenAPI document.
Authentication The type of authentication to use for the integrations that use this connected system.
  • None: No specific authentication will be applied. You can provide custom authentication values in the integration's URL, parameters, or headers as required by the external system. Client certificate/mutual SSL authentication can be enabled by uploading client certificates in the Admin Console.
  • Basic: HTTP Basic authentication will be applied using the username and password provided.
  • OAuth 2.0: OAuth 2.0 authorization will be applied using the fields provided.
  • API Key: API key authentication allows you to securely store and pass an API key as either an http header or a query parameter.

Amazon Machine Learning

Easily connect to Amazon Web Services Machine Learning models to make predictions based on your data. Pick from available models and quickly map your data to build smarter applications for fraud detection, demand forecasting, targeted marketing, and more.

In addition to the common properties, an Amazon Machine Learning connected system has the following properties:

Field Description
Access Key Id Amazon uses a custom authentication schema that requires a special signing of the request using an access key, secret access key, and region. The access key id, along with an associated secret access key and region, are used to access Amazon services. Your access key id is the first piece of a set of programmatic credentials that can be generated for Amazon Identity and Access Management (IAM) user accounts.
Secret Access Key Your secret access key is the second piece of a set of programmatic credentials that can be generated for Amazon IAM user accounts.
Region Your region is the third piece of a set of programmatic credentials that can be generated for Amazon IAM user accounts.
Test Connection Uses the provided configuration information to connect to the target system. Returns success or error with details.

Microsoft Azure LUIS

Easily connect to Microsoft Azure Language Understanding (LUIS) machine learning models to understand what your users want in their own words. Integrations using this connected system will allow you to provide a user's input and LUIS will extract the intent and associated entities represented by the text.

In addition to the common properties, a Microsoft Azure LUIS connected system has the following properties:

Field Description
Subscription Key Subscription keys can been found in your Azure portal or in the “Publish” tab of the LUIS application. In the “Publish” tab, any key that is NOT labeled “Starter_Key” is a subscription key.
Subscription Key Region Region for LUIS Endpoint API. The region in which you publish your LUIS app corresponds to the region or location you specify in the Azure portal when you create an Azure LUIS endpoint key. When you publish an app, LUIS automatically generates an endpoint URL for the region associated with the key.
Starter Key This key can be found in the “Publish” tab of the LUIS portal for any of your LUIS apps and is called “Starter_Key”. This key is also referred to as the “Programmatic Key” in the Microsoft documentation.
Starter Key Region Region for LUIS Programmatic API. The authoring region app can only be published to a corresponding publish region.
Bing Spell Check Subscription Key Specifying this value allows you to run spell check on query text before sending it through the model. You can create a Bing Spell Check key in your Azure portal.
Use Staging? When checked, Appian will query your staging model. Use this feature to validate changes to your model. You should always uncheck this box in production using import customization.
Test Connection Uses the provided configuration information to connect to the target system. Returns success or error with details.

Google Cloud Natural Language

Easily connect to Google Cloud Natural Language machine learning models. Extract structure and information from text, including both sentiment and key subjects such as people, places, and events. Integrations using this connected system will allow you to provide a text value and select which types of information to analyze.

In addition to the common properties, a Google Cloud Natural Language connected system has the following properties:

Field Description
API Key Google API keys can be generated from Google’s cloud console. Be sure to restrict the API key to Google Natural Language.
Test Connection Uses the provided configuration information to connect to the target system. Returns success or error with details.

Google Cloud Vision

Easily connect to Google Cloud Vision machine learning models. Extract structure and information from images, including logos, landmarks, and text (Optical Character Recognition). Provide an image and select different dimensions to analyze.

In addition to the common properties, a Google Cloud Vision connected system has the following properties:

Field Description
API Key Google API keys can be generated from Google’s cloud console. Be sure to restrict the API key to Google Cloud Vision.
Test Connection Uses the provided configuration information to connect to the target system. Returns success or error with details.

FEEDBACK