Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.superbryn.ai/docs/llms.txt

Use this file to discover all available pages before exploring further.

​
When to create manually

While AI generation covers most use cases, you may want to create scenarios manually for:
  • Specific edge cases you’ve seen in production
  • Regression tests for known bugs
  • Custom test cases that don’t fit standard ring patterns
  • Fine-tuned scenarios with precise expected outcomes

​
Creating a scenario

  1. Go to the Evaluators page
  2. Click Add Scenario
  3. Fill in the fields:
FieldDescription
RingWhich ring this scenario belongs to
IntentWhat the caller wants
User PerspectiveSecond-person description of the caller’s behavior
Expected OutcomeStep-by-step expected agent behavior
PathWhich call flow path this exercises
Variant Type(Ring 2) Which policy rule is being tested
Behavior ModifiersOptional: language, interruption timing, noise, etc.

​
Editing scenarios

Click on any scenario in the evaluator list to open the Scenario Detail Panel. All fields are editable. Changes are saved when you confirm.

​
Tips for writing good scenarios

User Perspective - Write in second person (“You call to…”). Be specific about what the caller says and does. Include realistic details from your domain. Expected Outcome - List each step the agent should take, in order. Use bullet points or numbered steps. Be specific enough to evaluate pass/fail. Intent - Keep it concise. This is the one-line summary of what the caller wants.