On this page
Build a branching support agent
A customer calls with an invoice question or refund request. The agent needs to verify account access, read the account and choose the right next step. If it cannot resolve the request, it should offer a specialist.
Build those decisions in Studio using conversation nodes, an HTTP action and a reusable refund-review component. The example records are fictional; your team supplies the account service and business rules.
Map the decisions
For the business access check below, your endpoint decides whether the caller may see the record. If you need verification through a private telephone keypad, use the private verification workflow instead of collecting sensitive answers in the conversation.
| Step | What you configure | What to check |
|---|---|---|
| Identify caller | Ask for the account reference and explain the purpose | Do not disclose account details before verification |
| Verify account | The verification conversation and confirmed-identity transition | Test refusal, wrong-person and ambiguous answers; enforce access checks in the account service |
| Look up account | A function node bound to your HTTP action | Validate the input and response schemas; define timeout and uncertainty behavior |
| Route request | A branch with conditions on variables | Make the default path explicit; missing variables must have a defined outcome |
| Refund review | A reusable component | Review eligibility before promising an outcome |
| Explain invoice | Instructions using the returned account facts and approved policy | Answer from available facts; escalate when they are insufficient |
| Transfer | A transfer node and a destination | Test unavailable or failed transfers as well as successful ones |
| End | The terminal node | Decide what completion means for this business workflow |
Define the verification and refund rules before connecting the nodes. Use conversation conditions for answers such as “I want a refund,” and equation conditions for returned values such as eligibility. Test ambiguous answers and missing values.
Bind an external action
- Open Settings → Actions and define
lookup_account. Supply the HTTPS endpoint, method, input/output schemas, timeout and credential reference. - Add a function node and select that action. Enable Wait for result when the next decision requires its response.
- Choose whether to speak during execution. A short status message can explain a pause without claiming that the request already succeeded.
- Configure response variables and the transitions that consume them. Treat an empty result, an error and an uncertain outcome as separate cases.
- Use a mock response in a simulation. Verify the real endpoint separately with an authorized integration run.
Credentials belong in secrets, not in prompts. Your endpoint still owns authentication, input validation and business authorization. For writes, it must honor the action's idempotency contract. Reconcile an uncertain write with the business service before retrying it.
For API configuration, see HTTP actions and draft edits. Review attempts and delivery results through action receipts.
Reuse a subflow
Move the refund review into a component with its own entry and return point. The main graph calls it when the customer requests a refund. Test eligibility, confirmation and rejection inside the component, then check how each result returns to the main flow.
Studio and the draft API edit the same definition. Publishing pins the main flow and its components together, so each run uses the component version you tested.
Ground answers in documents
For policy questions, upload your approved sources through Knowledge and documents, wait for the index to be ready and pin the source documents on the draft. Publish that draft to make the knowledge configuration part of the version.
Keep account facts and policy retrieval separate: the account action reads the customer's record; knowledge retrieval finds supporting policy. Include missing or conflicting policy evidence in your test cases.
Validate, test and release
A candidate published with live: false can only run with test: true. To use the reviewed behavior for regular calls or tasks, publish it with live: true; moving a test version to the head does not change that restriction. Publication and recovery covers the exact requests.
- Save the draft and run validation. Fix errors; review warnings against the intended behavior.
- Publish a test version with
live: false. Run the regression suite against that exact version. - Inspect failed assertions and tool behavior. Edit the draft, publish another test version and rerun the same suite.
- Refresh and validate the reviewed draft, then publish it explicitly with
live: truefor regular calls or tasks. Verify its returned digest and test flag. Existing runs keep their original publication; assign phone routes separately. - Review a real run using the run investigation guide. Test real business endpoints, carrier connectivity and audio separately.
Each run records its publication. When a production issue appears, open that version to inspect the flow and component that executed.


