Post-deployment tasks
Complete these tasks after a GenAI local package is deployed to a target tenant:
- Validate target tenant data
- Validate the Created By field
- Test imported resources
Validate target tenant data
After deployment, verify that:
- All selected resources, including tools, agents, skills, extensions, focus modes, guardrails, conversation starters, and prompts, are displayed in the Factory and Prompt Catalog.
- Each resource retains the enabled or disabled status from the source tenant.
- The deployment does not create duplicate resources.
Note: Conversation starters can create duplicate entries on the target tenant. Review the Conversation Starters list and remove duplicate entries manually.
- Custom security roles (SRs) and functional security roles (FSRs) are present on the target tenant.
- MCP tools that are assigned to agents are present and function correctly.
- Tools that invoke custom API suites resolve correctly through the target tenant API Gateway.
- Prompt Catalog entries are displayed with the correct settings. Re-create attachments and variables that the deployment did not transfer.
Validate the Created By field
Verify that, for imported agents, the field is set to System both in the GenAI Factory and in the API Gateway. This confirms the resource was deployed through a local package and not created manually.
Test imported resources
- 1. Open .
- Submit prompts that match the imported agents' utterances.
- Verify that:
- The Supervisor Agent routes requests to the correct imported agent.
- Tools invoke and return valid API responses.
- Skills, extensions, and focus modes behave as expected.
- Guardrails block expected content.
- Conversation starters are displayed in the GenAI Assistant.