Create, edit, and organize test cases and suites for your bots
No test suites yet
Create a suite to group test cases together for organized execution
Test Cases
No test cases in this suite yet.
Use the checkboxes in the test case list below to add them.
Use the checkboxes in the test case list below to add more.
Child Suites
Versions
Create Test Suite
Edit Suite
| { if (!selectedTestCaseIds.includes(tc.id)) selectedTestCaseIds.push(tc.id) }) : category.testCases.forEach(tc => { const i = selectedTestCaseIds.indexOf(tc.id); if (i >= 0) selectedTestCaseIds.splice(i, 1) })"> | Case ID | Name | Version | Bot | Channel | Intent | Labels | Policies | Type | Steps | Actions |
|---|---|---|---|---|---|---|---|---|---|---|---|
|
|
Unassigned | 📞 Voice 💬 Chat |
+
|
+
-
|
~ |
No test cases found. Create or generate some!
No tests match your filter
Import Built-in Test Cases
Add prebuilt test cases from industry-standard frameworks to your bot.
Model theft, data poisoning, evasion, agents
Hate speech, self-harm, WMD, cybercrime
Prompt injection, info disclosure, agency (2025)
BOLA, auth, SSRF, business logic
Jailbreaks, encoding, roleplay, toxicity
Judging Mode
Response time SLOs
Greetings, help, thanks, goodbye
Unknown queries, invalid inputs
Graceful degradation, escalation
Conversation flow, context
Long inputs, Unicode, edge cases
Frustrated, happy, urgent users
Simple language, typo tolerance
Judging Mode
Performance tests ship with keyword patterns. AI Judge is recommended for more accurate results.
Risk management, transparency, fairness
PII protection, abuse, moderation
About Compliance Tests
Compliance tests validate your bot against regulatory and governance standards including NIST AI Risk Management Framework, PII handling, and content moderation policies. All compliance tests use AI Judge for contextual evaluation.
Import Test Cases
Import test cases from a YAML or JSON file.
Importing to bot: Importing as unassigned test cases (no bot selected)
Click or drag a YAML/JSON file here
Supports .yaml, .yml, .json
Assign to Bot
Currently assigned to: This test case is currently unassigned.
Applied Policies
+ inherits from bot defaultsNo policies applied
Judge Override
Inherits from bot configuration
Intent
Judgment Type
Input variation enabled
Latency SLO
From Policies
Test Steps
User Input:
User Input:
No user input
Expected Response Contains:
Judging:
Judge Hint:
Assertions
Minimum Scores
Task
Policy
Latency
Create Test Case
Manually create a test case with specific inputs and expected responses.
Edit Test Case
Assign Judges to Test Case
Select judges for this test case. Leave all unchecked to inherit from bot.
Loading judges...
Tip: If no judges are selected, this test case will use the bot's default judges.
Assign Policies to Test Case
Select policies to apply to this test case.
Task
Safety
Compliance
UX
No policies available.
Create policies