DESIGNER GUIDE
Test and troubleshoot your solution
Test workers directly, run complete customer journeys, investigate incomplete configuration and confirm that a project is ready before publishing or deployment.
Scope: draft worker or Flow project
Test behaviour, not only appearance
Three levels of testing
Each level confirms something the others cannot.
| Level | What it confirms |
|---|---|
| Direct worker test | The worker interprets the request and returns the expected response contract |
| Integrated flow test | The flow passes data to the worker or step and handles the result correctly |
| End-to-end Run Demo | The complete user journey behaves correctly from the first interaction to completion |
On this page
Before you begin
Prepare the draft, the data and the expected results before running the first test.
- Save the current draft.
- Use fictional test data.
- Define the expected outcome before each test.
- Prepare happy-path, incomplete and invalid examples.
- Identify every important branch.
- Confirm the expected worker-response structure.
- Reset conversational state between independent scenarios.
- Record the project version being tested.
- Avoid testing production actions from a draft environment.
1. Prepare a test matrix
A short matrix keeps coverage honest and repeatable across design changes.
| Scenario | Example | Expected result |
|---|---|---|
| Clear supported request | “Where is my package?” | Late-delivery or order-tracking intent is selected |
| Alternative wording | “My parcel should have arrived yesterday” | The same approved process is selected |
| Missing information | “My delivery is late” without an order number | The solution asks for the required information |
| Ambiguous request | “There is a problem with my order” | The worker requests clarification or returns uncertainty |
| Unsupported request | A question outside customer support | Approved fallback, main menu or human support |
| Invalid input | An incorrectly formatted order reference | Controlled validation message |
| Sensitive action | A request requiring approval | Human approval or approved escalation path |
| Adversarial input | An attempt to override the worker’s role | Installed guardrails and instructions prevent unsafe behaviour |
2. Test an Agentic Worker directly
Start with the worker in isolation so interpretation problems are not confused with flow problems.
- Reset chat
- Clears the current test conversation so an independent scenario does not inherit earlier context.
- Suggested messages
- Quick starting examples. They are not a complete test set.
- Message input
- Accepts the test request.
- The user enters an open customer question.
- The worker identifies a likely late-delivery intent.
- The response includes structured routing information.
- The result names the process intended to handle the request.
- The exact fields shown are specific to this demonstration.
- 1Open the worker in Composer.
- 2Select the Test panel.
- 3Select Reset chat.
- 4Enter one test message.
- 5Compare the returned result with the expected contract.
- 6Confirm the detected intent.
- 7Confirm extracted values and uncertainty handling.
- 8Confirm that the worker does not attempt the deterministic resolution itself.
- 9Repeat the test with alternative wording.
- 10Test a missing-information request.
- 11Test an unsupported request.
- 12Reset the chat before each independent scenario.
3. Test the complete flow in Run Demo
Run Demo tests the current flow through the same structured interactions the user will experience.
- Initial content and instructions
- Buttons and structured choices
- Open-text input
- Required fields
- Validation messages
- Worker routing
- Called flows
- Completion messages
- Fallback behaviour
- 1Save the current draft.
- 2Open Run Demo.
- 3Begin with the main happy path.
- 4Complete each displayed input.
- 5Confirm the expected next step appears.
- 6Complete the entire journey.
- 7Restart the interaction.
- 8Test an alternative branch.
- 9Test missing or invalid data.
- 10Test an open-text request that invokes the worker.
- 11Confirm that the correct deterministic flow starts.
- 12Repeat until every important route has been covered.
4. Test worker-to-flow routing
Interpretation, integration and the end-to-end process each need their own confirmation.
1. Test the worker directly
↓
2. Test the Agentic Worker step in the flow
↓
3. Test the complete process from Run Demo- Stage 1: Worker
- Confirm that the request produces the expected intent and structured fields.
- Stage 2: Flow integration
- Confirm that the Agentic Worker step passes the correct input and that the following branch receives the expected result.
- Stage 3: End-to-end process
- Confirm that the customer request starts the intended deterministic resolution flow and completes successfully.
- 1Enter “Where is my package?” in the worker Test panel.
- 2Confirm the worker returns a late-delivery classification.
- 3Enter equivalent wording in the customer-facing Run Demo.
- 4Confirm the main flow invokes the router.
- 5Confirm the late-delivery branch starts.
- 6Confirm the deterministic resolution requests the required information.
5. Investigate a project issue
An issue count is a signal to review required configuration and connections.
- 1Save the current project.
- 2Note the displayed issue count.
- 3Inspect recently added or changed steps.
- 4Confirm that every required field has a value.
- 5Confirm that referenced workers, flows and tools are available.
- 6Check that process steps have the required incoming and outgoing connections.
- 7Check branch outcomes and destinations.
- 8Confirm that expected variables are present.
- 9Run Demo again.
- 10Continue to the readiness check after the issue is resolved.
| Check | Example |
|---|---|
| Required resource | Agentic Worker step has an available worker selected |
| Required destination | Go to Flow points to an available flow |
| Required capability | MCP, HTTP or extension resource is available |
| Connection | The step is linked to the intended next route |
| Branch | Every expected result has a controlled destination |
| Input | Required data is passed into the step |
| Output | The following step expects the fields actually returned |
6. Confirm readiness before shipping
The readiness check confirms when Designer found no blocking configuration issues.
- A successful result states that Designer found no issues.
- The summary may show cards and step counts.
- Let’s go continues toward publishing or deployment.
- A successful technical result should be combined with business testing.
- 1Resolve the current issue count.
- 2Save the draft.
- 3Run each priority test.
- 4Record the test result.
- 5Continue toward deployment.
- 6Review the readiness result.
- 7Return to the Editor if a blocking issue remains.
- 8Continue only after both technical validation and business review succeed.
The next stage is covered in the Publishing and deployment guide.
7. Retest after a change
A correction is only complete when the failed scenario passes and nearby scenarios still pass.
Reproduce
↓
Change one thing
↓
Save
↓
Run the failed scenario
↓
Run a nearby successful scenario
↓
Run validation
↓
Save as version- 1Reproduce the problem before changing the design.
- 2Make one focused correction.
- 3Save the draft.
- 4Repeat the failed test.
- 5Repeat at least one related passing test.
- 6Check that the correction did not break another branch.
- 7Run the readiness check.
- 8Save a meaningful version when the correction is accepted.
Testing checklist
Confirm this list before a solution moves toward publishing or customer acceptance.
- Expected results were defined before testing.
- Only fictional test data was used.
- The worker was tested directly.
- Independent worker tests used Reset chat.
- The happy path completed successfully.
- Alternative wording was tested.
- Missing information was tested.
- Invalid input was tested.
- Ambiguous requests were tested.
- Unsupported requests followed the approved fallback.
- Important branch boundaries were tested.
- Worker output matched the flow contract.
- Called flows started correctly.
- Human approval paths were considered.
- The visible issue count was resolved.
- A readiness check completed successfully.
- Business acceptance was recorded separately.
- The accepted project state was saved as a version.
Testing troubleshooting
Frequent testing problems and the practical action that resolves them.
| Problem | Likely cause | Action |
|---|---|---|
| A direct worker test is influenced by an earlier message | The test conversation still contains previous context | Select Reset chat before the independent scenario |
| The worker chooses the wrong intent | Instructions or intent boundaries are unclear | Refine the definitions and test alternative wording |
| The worker returns extra prose around structured output | The response contract is too loose | Require only the specified structure and retest |
| The worker is correct but the wrong flow branch starts | The flow checks the wrong field or value | Compare the worker output with the branch condition |
| Run Demo waits without continuing | The current step requires user input or a button action | Complete the displayed interaction and verify the intended next step |
| A step has no available resource | The required worker, flow or integration is unavailable | Configure, publish or expose the referenced resource |
| The toolbar shows an issue | One or more required settings or connections are incomplete | Review recently changed steps and required fields |
| The readiness check fails | A blocking project issue remains | Return to the Editor and correct the reported configuration |
| The readiness check passes but users receive the wrong outcome | Technical validation did not cover the business rule | Expand the test matrix and correct the process design |
| A correction breaks another branch | Shared variables or connections changed | Retest nearby paths and restore the intended mappings |
| Test results vary too much | Worker instructions or model settings allow excessive variability | Tighten the contract and review model configuration |
| Real systems were affected during a draft test | Production resources were connected to the test process | Stop testing, review integration scope and use a safe test environment |
