Skip to main content

​
What is Infrastructure Suite?

Infrastructure Suite is a collection of pre-built test scenarios designed to validate the reliability of your voice AI infrastructure. With one click, you can add 18+ ready-made test cases to your project that cover latency, audio quality, interruption handling, language support, and edge cases. Rather than building infrastructure tests from scratch, Infrastructure Suite provides “batteries included” testing that works across all voice stacks — LiveKit, Pipecat, VAPI, Retell, and custom real-time setups.

​
Why Use Infrastructure Suite?

Save Time

Skip the manual work of creating infrastructure test scenarios. Add 18+ pre-configured tests instantly.

Ensure Reliability

Validate that your infrastructure handles real-world challenges like interruptions, background noise, and latency spikes.

Continuous Validation

Run Infrastructure Suite tests regularly in CI/CD to catch infrastructure degradation before it impacts users.

Best Practices

Leverage test scenarios built from common infrastructure failure patterns observed across production systems.

​
What’s Included

Infrastructure Suite includes 18+ test scenarios covering hold tests, audio quality under various conditions, interruption handling, language support, and edge cases like packet loss and rapid-fire phrases. Complete list of Infrastructure Suite test scenarios

​
How to Add Infrastructure Suite

1

Navigate to Evaluators

In the Cekura dashboard, go to your agent’s Evaluators tab in the sidebar.
2

Locate Infrastructure Suite Folder

In the Folders section, you’ll see the Infrastructure Suite folder labeled as “Predefined infrastructure tests”.Infrastructure Suite folder in the Evaluators sidebar
3

Preview Test Scenarios

Click Click to Preview to view all available test scenarios and their details.
4

Add to Your Project

Click + Add to my Project in the top-right corner. This copies all scenarios to your project with the appropriate personality and metric mappings.
Infrastructure Suite scenarios are copied (not linked) to your project. You can modify them as needed without affecting the original templates. Global personalities are reused, while others are created as org-scoped copies.
A Rubric rule will be added to your project: AI Interrupting user = 0. This rule affects pass/fail on runs with interruption-related scenarios. You can remove it later from Rubric settings if not relevant to your use case.

​
Running Infrastructure Tests

Once added, select the scenarios from your Evaluators tab and click Run. Use the frequency parameter for load testing — see Load Testing for details.

​
Integration with CI/CD

Infrastructure Suite tests are ideal for continuous integration workflows. Use the Cekura GitHub Actions to run infrastructure validation automatically: 
- uses: cekura-ai/cekura-github-actions@v1.0.0
  with:
    agent_id: ${{ vars.AGENT_ID }}
    tags: "infrastructure-suite"
    api_key: ${{ secrets.CEKURA_API_KEY }}
Tag your Infrastructure Suite scenarios with infrastructure-suite to run them as a group in CI/CD pipelines. Learn more about GitHub Actions CI/CD.