> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://poly-ai.ferndocs.com/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://poly-ai.ferndocs.com/_mcp/server.

# Test suite

> Save conversations as test cases and re-run them to check for regressions.

The **Test suite** lets you save real conversations as test cases and re-run them against Draft or Sandbox to check for behavioral drift. Each case captures the full conversation — user messages, agent replies, and tool calls — so you can verify the agent still handles the scenario correctly after changes.

## Create a test case from a conversation

#### Open a conversation in Conversation Review

Go to **Analytics > Conversations** and select a conversation. In the side panel, click the **Add test** button.

![Add test from Conversation Review](/_fern-files/_img/e133d9a977fa63b05adabcdc30eff46509d7baf5461a8e8cfcf9f52cce21b79c.webp)

#### Name the test case

In the **Create test case** modal, give the case a descriptive name (for example, "Sunday Opening Test") and click **Create**.

![Create test case modal](/_fern-files/_img/c1c0dee12be373c52a1aca05c3b55968b10470e79eafccca18c0ac3533146c2c.webp)

#### Review the test case

The test case opens with the full conversation transcript. Edit assertions and user input to define what counts as a successful run.

![Test case view](/_fern-files/_img/db1f840f9f3d358b84ff22d91ab7db25efe4d010a937877947756c3d64b5beb6.webp)

## Edit test case parameters

Each test case stores the function call values from the original conversation. You can edit these to test variations of the same scenario without creating a new case.

1. Open the test case from **Test suite**.
2. Select the parameters you want to modify.
3. Adjust values to simulate a different scenario — for example, change a date, customer ID, or location.
4. Save the case.

> **Tip**
>
> Editing parameters is useful for testing edge cases. For example, duplicate a booking test case and change the party size to test large-group handling.

## Best practices

* **Save tests from real conversations** — cases from [Conversation Review](/agent-studio/conversations/review) capture real-world scenarios your agent has already encountered.
* **Name cases descriptively** — use names that describe the scenario, not the expected outcome (for example, "Sunday Opening Test" rather than "Test 1").
* **Cover edge cases** — duplicate a case and tweak parameters to test boundary conditions without re-recording a conversation.