Simulate Users

Use UserSimulator to drive multi-turn tests with LLM-generated user inputs.

1. Configure a generator

To get started, you need to provide the LLM that will power the simulator. UserSimulator uses a generator to produce each user turn, so the same model you use for your checks can also drive realistic user behavior.

UserSimulator uses an LLM to generate realistic user messages. Set a default generator once, or pass one inline.

def support_agent(message: str) -> str:
    """Stub support agent for demonstration."""
    return "I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with?"

from giskard.checks import set_default_generator
from giskard.agents.generators import Generator

set_default_generator(Generator(model="azure_ai/gpt-4.1-nano"))

2. Create a UserSimulator with a persona

With the generator configured, we can now define who the simulated user is. The persona field acts as a system prompt for the simulator — it describes the user’s role, goal, and stopping condition. The more specific you are, the more deterministic and useful the generated conversation will be.

from giskard.checks.generators.user import UserSimulator

customer = UserSimulator(
    persona="""
    You are a customer trying to track a delayed order.
    - Start by asking about order #98765
    - Provide your name (Alex) when asked
    - Accept any resolution the support agent offers
    - Stop when the agent confirms a solution
    """,
    max_steps=8,
)

max_steps limits how many turns the simulator will generate before stopping.

3. Use the simulator as inputs in .interact()

Now we’ll wire the simulator into the scenario. Passing the UserSimulator as inputs tells the scenario to call it on each turn rather than using a fixed string — the scenario handles the loop automatically up to max_steps.

Pass the UserSimulator instance as the inputs argument. The scenario will call it repeatedly to generate each user turn.

from giskard.checks import Scenario, FnCheck

scenario = (
    Scenario("order_tracking")
    .interact(
        inputs=customer,
        outputs=lambda inputs: support_agent(inputs),
    )
    .check(
        FnCheck(fn=
            lambda trace: any(
                word in trace.last.outputs.lower()
                for word in ["resolved", "refund", "replacement", "shipped"]
            ),
            name="resolution_offered",
        )
    )
)

4. Run the scenario and inspect the trace

With the scenario built, run it and iterate over the trace to see the full conversation the simulator generated. This is especially useful when debugging a failing check — you can see exactly what the simulated user said at each step.

import asyncio
result = asyncio.run(scenario.run())

# Print every turn
for turn in result.final_trace.interactions:
    print(f"User:  {turn.inputs}")
    print(f"Agent: {turn.outputs}")
    print()

Output

User: Hello, I would like to inquire about the status of my order #98765. It was supposed to arrive last week, but I haven’t received it yet. Can you please provide an update? Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with?

User: Hello, I can help you with that. Could you please confirm the name associated with the order? Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with?

User: Hello, I would like to confirm if my order #98765 has been delivered yet. Thank you. Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with?

User: Thank you for the update. Could you please confirm if the delivery has been completed now? Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with?

User: Hello, I appreciate the update. Could you please confirm whether my order #98765 has been delivered now? Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with?

User: Hello, I wanted to check if my order #98765 has been delivered today. Thank you. Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with?

User: Hello, I understand that my order #98765 is scheduled to arrive tomorrow. Since I haven’t received it yet, could you please confirm whether it has been delivered today? Thank you. Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with?

User: Hello, I understand that my order #98765 is scheduled to arrive tomorrow. Since I haven’t received it yet, could you please confirm whether it has been delivered today? Thank you. Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with?

5. Check goal_reached from simulator metadata

After the scenario finishes, the simulator writes a LLMGeneratorOutput into the last interaction’s metadata. This tells you whether the user’s stated goal was achieved, a stronger signal than just checking whether the scenario passed its checks, because it reflects the simulator’s own evaluation of the conversation outcome.

from giskard.checks.generators.base import LLMGeneratorOutput

last = result.final_trace.last
simulator_output = last.metadata.get("simulator_output")

if isinstance(simulator_output, LLMGeneratorOutput):
    print(f"Goal reached: {simulator_output.goal_reached}")
    print(f"Message: {simulator_output.message}")

Use goal_reached as an additional assertion:

if simulator_output and not simulator_output.goal_reached:
    print(f"Goal not reached: {simulator_output.message}")
else:
    print("Goal reached or no simulator output")

Output

Goal reached or no simulator output

6. Swap personas for A/B testing

With a single persona working, we can now run the same agent against multiple user types simultaneously. Each persona exercises a different interaction style, and running them concurrently with asyncio.gather means you get results for all three in roughly the time it takes to complete one.

Run the same agent against multiple user types to surface persona-specific failures.

import asyncio

personas = [
    (
        "impatient",
        "You are impatient. Keep messages short. Escalate quickly if not helped.",
    ),
    (
        "detailed",
        "You are thorough. Ask many follow-up questions before accepting any solution.",
    ),
    (
        "confused",
        "You are unsure what you need. Describe symptoms, not the actual problem.",
    ),
]


async def run_persona(name, instructions):
    sim = UserSimulator(persona=instructions, max_steps=6)
    scenario = Scenario(name).interact(
        inputs=sim,
        outputs=lambda inputs: support_agent(inputs),
    )
    return name, await scenario.run()


results = asyncio.run(asyncio.gather(*[run_persona(n, i) for n, i in personas]))

for name, result in results:
    print(f"{name}: {'PASSED' if result.passed else 'FAILED'}")

Output

impatient: PASSED detailed: PASSED confused: PASSED

Custom trace formatting

By default the trace prints interactions as raw inputs and outputs. You can write a simple formatting function to produce a human-readable transcript — for example, to log a simulated conversation or include it in a test failure message. For a subclass of Trace, Rich rendering, and how that interacts with print_report(), see Custom trace types.

def format_transcript(trace) -> str:
    """Format a trace as a human-readable chat transcript."""
    lines = []
    for turn in trace.interactions:
        lines.append(f"User:  {turn.inputs}")
        lines.append(f"Agent: {turn.outputs}")
    return "\n".join(lines)


result = await (
    Scenario("chat_trace_demo")
    .interact(
        inputs=customer,
        outputs=lambda inputs: support_agent(inputs),
    )
    .run()
)

print(format_transcript(result.final_trace))

Output

User: Hello, I would like to check the status of my order #98765. It appears to be delayed. Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with? User: Thank you for the update. Could you please confirm the estimated delivery date? My name is Alex. Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with? User: Thank you for confirming the arrival date. Could you please advise on the next steps to receive my order? Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with? User: Could you please confirm if there are any updates regarding the delivery schedule? I appreciate your assistance. Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with? User: Thank you for the information. Could you please confirm the expected delivery date for my order? My name is Alex. Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with? User: Thank you for the update. Could you please confirm the exact delivery time today? Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with? User: Could you please confirm if there are any updates regarding the delivery schedule today? Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with? User: Hi, thank you for the update. Could you please confirm the exact delivery time today for my order? My name is Alex. Agent: I have located your order #98765. It is currently in transit and will arrive tomorrow. Is there anything else I can help you with?

Next steps

• Generators reference — full UserSimulator parameter reference
• Multi-turn testing tutorial — multi-turn scenario basics
• Debug with Spy — inspect what happens inside each interaction