Scenarios
Last updated
Last updated
A scenario in OpenDialog encompasses the set of functionalities that make up your AI application.
Each workspace holds one or more scenarios. You can access the Scenario overview through the left-hand panel menu, by selecting 'Scenarios' and then 'Overview'.
You can also do so by accessing the Scenario overview through the dedicated button 'Manage scenarios'.
Create a new scenario using the main area's 'Create Scenario' button.
Scenarios are by default, created in draft mode. To use in a live environment, set the scenario to active by toggling the switch in the bottom right corner of the scenario tile (example shown in top left corner of scenario list), and publish it.
Click the vertical dots icon in the top right corner of a scenario tile to access the duplicate, delete and export functionality.
When duplicating a scenario the duplicated scenario will be added to the workspace as the last one in the list. You may have to refresh your browser for it to display.
When you export a scenario, an email is sent to your registered email address in OpenDialog. The scenario is contained in an attached .json file, which can then be imported into another workspace.
Duplicating and exporting may take a few minutes; you will be emailed when the resulting file is ready.
To import a scenario, you will need access to a .json file that you will upload during import. For example, you may have received a .json file as an export from another workspace or as an email attachment.
Use the Import Scenario button at the top of the main area and provide the .json file. Note that there is a 10 Mb file size limit on scenario .json files.
Scenarios can be in either an "Active" or in "Draft" mode.
Active scenarios are available to be used in chat interfaces, while draft scenarios are not considered.
You can edit the Scenario name, description, set a preferred interpreter (which will act as the default interpreter throughout the scenario) and also set conditions. Setting a condition on a scenario is relevant in a multi-agent architecture, where conditions on scenarios identify a scenario for consideration.
