Builder

Build the conversation your agent will have, then run it on a real call.

The builder is a drag-and-drop canvas. The graph you draw is the agent — there is no separate configuration behind it, and nothing to deploy. Change the graph, save, publish, and the next caller reaches the new version.

Core components

  • Node — one step in a call: what the agent opens with, a question it asks, a system it calls, how it ends. Nodes catalogues all ten types.
  • Edge — the rule that moves the call from one node to the next. An edge fires when the model judges the caller's goal met, when a keypad digit is pressed, or when a condition on gathered data holds.
  • Start Call — the single entry point. Every workflow has exactly one, it cannot be deleted, and nothing can point an edge into it.
  • Draft — your unpublished edits. Real callers never reach a draft.
  • Run — one execution of the graph against one caller. Runs carry the transcript, the recording, the cost and the outcome; the workflow itself carries none of those.
A workflow on the builder canvas: a Start Call node at the top, five Agent Nodes, a Global Node holding the persona, and three End Call nodes, joined by edges whose labels read good time, busy — callback, has a need, objection / hesitation, qualified, nurture, resolved — qualified, resolved — nurture, firm no, email + consent and declined.
A ten-node outbound agent on the canvas. Each pill on an edge is the condition that moves the conversation to the next node.

Building a workflow

A new workflow arrives with one node and no edges: Start Call. Everything else you add.

  1. Add a node from the node panel, or right-click the canvas and pick a type. Start Call is absent from both lists, because a second one is not allowed.
  2. Drag from a node's handle to another node to draw an edge, then label it with the condition that should fire it.
  3. Open a node to write its prompt and attach whatever it needs — tools, documents, variables to extract.
  4. Save. Saving records a draft; it does not change what callers hear.

Select a node and delete it from its toolbar, or with the Backspace key. Start Call has neither affordance.

Routing

A node does nothing on its own. It runs when an edge leads into it, and it hands off when one of its outgoing edges resolves.

Most workflows are a chain of Agent Nodes, each with a focused prompt, joined by edges that describe the transition between them. Two nodes sit outside that chain:

  • A Global Node contributes persona and standing rules to every node that opts in. It is never a step, so no edge touches it.
  • An API Trigger node is an entry point from outside the call rather than a step within it.

Start Call may sit on the canvas with no outgoing edge at all — a one-node workflow is valid and publishable. That is what you get on creation, and it is why a brand-new workflow can be tested immediately.

Draft and published

Edits on the canvas are held as a draft until you publish. The published version keeps running for every real caller while you keep editing. Publishing promotes the draft and archives the version it replaces. See Versions.

Testing

The Test menu holds three ways to hear a change, and all three run your draft rather than the published version:

  • Web Call places a call over WebRTC from the browser. It opens a full-page run screen — microphone controls on one side, a live transcript building on the other — so the canvas is not on screen while the call is up.
  • Phone Call places an outbound call to a number you type.
  • Text chat runs the same graph as typed messages instead of speech.

The menu stays disabled while the workflow has unsaved changes or validation errors. Save and fix the errors first.

Next steps

GoalGuide
Choose the right stepNodes
Decide what a node saysPrompts
Pick how the agent soundsVoices
Edit safely while calls are liveVersions

Common questions

What is Woise?

Woise lets you build AI agents that answer and make phone calls and chat on your website. You lay out the conversation on a visual canvas, try it in your browser, then connect it to a phone number.

Do I need a phone number to start?

No. You can build and test an agent in the browser without one. When you are ready, connect a phone number and choose which agent answers it.

How does billing work?

You pay for the minutes your agents use, from your credit balance. There are no seat fees and no per-agent fees, so an agent that is not taking calls costs nothing.

Which languages can an agent speak?

Agents can listen and speak in more than 40 languages, including English, Spanish, Hindi, Tamil and Arabic. You pick the language and the voice for each agent.

Can an agent connect to my CRM or calendar?

Yes. An agent can look things up and make changes in 26 apps, such as Salesforce, HubSpot, Google Calendar, Slack and WhatsApp, while the caller is still on the line. You connect each app once.

Do I need engineers to build an agent?

No. Everything is built in a visual editor, so you can create and change an agent without writing code. If your team prefers code, there is also a REST API and an MCP server.

What happens to call recordings and transcripts?

Every call keeps a transcript and its outcome, and audio is only recorded if you turn it on. Only people in your account can see your calls, and your account lives in the region you chose when you signed up.

What do I have to set up or run?

None. We run the speech, the language models, the recordings and the phone connections, and scale them with your call volume. There is nothing for you to set up or keep running.

How do I reach the team?

Email contact@woise.ai, call or WhatsApp +91 83412 34080, or use the form on our contact page. A person reads every message.

Still stuck? Talk to our team.