-Ace
03/05/2026, 1:15 AMAI Evy
03/05/2026, 1:15 AMAI Evy
03/05/2026, 1:16 AMname (required) — Name of the test case
- user_prompt (required) — User prompt to simulate
- metrics (required) — Array of metric names to evaluate
- response_engine (required) — Response engine configuration
- dynamic_variables (optional) — Dynamic variables to inject
- tool_mocks (optional) — Mock tool calls for testing
- llm_model (optional) — LLM model to use (e.g., gpt-4.1, gpt-4.1-mini, claude-4.5-sonnet)
Example structure from a test case: [(3) ]()
json
{
"name": "Normal Flow - Printer Offline Issue",
"dynamic_variables": {
"assigned_tech": "John Martinez",
"ticket_number": "T20231222-0145"
},
"metrics": [
"Ticket created successfully",
"User received ticket number"
],
"user_prompt": "Hi, my name is Alex Johnson...",
"type": "simulation",
"tool_mocks": [],
"llm_model": "gpt-4.1-mini"
}AI Evy
03/05/2026, 1:16 AMAI Evy
03/05/2026, 1:16 AMAI Evy
03/05/2026, 1:16 AM