Step 10. Configure and publish the processes

1. 1To publish the following processes, click Actions in the row that contains the processes and select Publish.
• - Generic Executor
• - Azure OpenAI Function Calling
• - Salesforce Executor
• - Azure OpenAI Agent for Salesforce
• - Query Dynamics 365
2. 2Open the Prompt Azure OpenAI LLM process.
3. 3On the Input Fields tab of the Send Prompt step, enter the deployment ID and the API version of the Azure OpenAI LLM model in the deployment_id and api_version fields.
4. 4Save and publish the process.
5. 5Open the Query Snowflake process.
6. 6On the Assignments tab of the Set Default Database and Schema Values step, enter the database and schema values for Snowflake database in the database and schema fields.
7. 7Save and publish the process.
8. 8Open the Azure OpenAI Agent for Dynamics 365 and Snowflake process.
9. 9On the Assignments tab of the Snowflake LLM System Instructions step, enter the following field values:
• - Update information about Snowflake tables in the Table_Metadata field.
• - Update LLM system instruction in the System_Instruction field, if required.
• - Enter the service URL that you get after publishing the Query Snowflake process in the API_URL field.
10. 10On the Assignments tab of the Dynamics 365 LLM System Instructions step, enter the following field values:
• - Update LLM system instruction in the System_Instruction field, if required.
• - Enter the service URL that you get after publishing the Query Dynamics 365 process in the API_URL field.
11. 11Save and publish the process.
12. 12Open the Microsoft Teams AI Agent process.
13. 13On the Start tab of the Start step, select Allow anonymous access.
14. 14On the Assignments tab of the Set Flow Configurations step, update the function declaration in the Function_Declaration field and configure the prompt instructions by updating the Generation_Config field using the Expression Editor, as shown in the following sample code:
<root>
<temperature>0.2</temperature>
<maxOutputTokens>2500</maxOutputTokens>
<topP>0.8</topP>
</root>
For the Generation_Config field, enter values for the following properties:

Property	Description
temperature	Controls the randomness of the model's output. A lower value makes the output more deterministic, while a higher value increases randomness and creativity. For example, a temperature of 0.5 balances between deterministic and creative outputs.
maxOutputTokens	Defines the maximum number of tokens the model can generate in its response. Setting a limit ensures that the response is concise and fits within the desired length constraints.
topP	Determines the cumulative probability threshold for token selection. The model considers the smallest set of tokens whose cumulative probability meets or exceeds topP. For example, if topP is set to 0.1, the model considers only the top 10% most probable tokens at each step.

15. 15On the Assignments tab of the Get Data from Message step, you can optionally update the team ID value in the Team_ID field.
To find the team ID, perform the following steps:
1. aGo to your team listed on the left side of the Teams app.
2. bSelect More options > Get link to team as shown in the following image:

3. cCopy the group ID value as shown in the following image:

16. 16Save and publish the process.
17. 17On the Actions menu of the Microsoft Teams AI Agent process, click Properties Detail and copy the service URL.
You must copy the service URL to set up the webhook configuration in Microsoft Teams. For more information, see Step 11. Create webhook in Microsoft Teams.