The Engine Playground lets you validate how your AI Security Engine Profiles behave before applying them to production AI Interaction Policy rules. It provides a controlled simulation environment where you can test prompts against selected profiles and review detection results in real time.
In the Playground, you choose which profiles to activate, select a predefined testing scenario or create a custom one, and run the simulator. The system evaluates the prompt against the active profiles and shows whether any profile violations were triggered.
This feature helps you:
Validate detection logic before enforcement
Identify gaps in profile coverage
Reduce false positives
Fine-tune custom detectors and profile combinations
Because profiles are often reused across multiple policy rules, testing them in advance ensures that once they are enforced, they behave exactly as intended.
Use Case
As a security admin for Company ABC, you recently created a profile called Personal Identifier Protection to prevent employees from sharing sensitive personal data with AI tools.
Before applying the profile to a blocking rule in your AI Interaction Policy, you want to verify that it correctly detects:
SSNs
Email addresses
Customer account numbers (custom regex detector)
You navigate to the Engine Playground and enable only the Personal Identifier Protection profile.
Next, you select the predefined Personal Information scenario and run the simulator. The system evaluates the prompts in the scenario against your profile and shows whether violations were triggered.
You notice that SSNs and email addresses are correctly detected, but your internal account number pattern is not triggering.
You update the custom regex detector, return to the Playground, and run the simulation again. This time, the violation is correctly identified.
Now you are confident that when you apply the profile to a blocking rule, it will behave as expected in production.
Test Profiles in the Playground
Use the Engine Playground to simulate prompts against selected profiles and review the detection results.
To test profiles in the Engine Playground:
From the navigation menu, select Security > AI Security > Playground.
In the Select Engine Profiles to Activate panel on the left, enable the profiles you want to test.
In the Session area, select the scenario you want to use:
Choose a predefined scenario, such as:
Personal Information
Jailbreak & Prompt Inection
API Keys & Secrets in Code
Or select Custom Session to create your own test prompts
If using a custom session, enter your prompt in the User Prompt field
Click Run Simulator.
Review the simulation results.
The system shows how many profiles the prompt was evaluated against. You see whether any Profile Violations were triggered. If violations occur, you can review which profiles were matched
Adjust profiles or prompts as needed and run the simulator again to validate your changes:
Modify detectors or profile logic
Enable or disable additional profiles
Update your test prompts