Skip to main content

​
The idea

You do not start from a blank YAML file. You talk to a builder agent in the left-hand chat, tell it what you want, and it authors the playbook. The result appears live in the Editor tab on the right, where you can refine it in chat or edit it by hand. ▶Open the PlaygroundEverything below happens at superdialog.unpod.ai/playground. A guest trial lets you build a playbook before creating an account.→

​
Step 1 - Describe what you want

SuperDialog landing composer: 'Turn a prompt into a conversation agent', a text box reading 'Describe the conversation you want to build - e.g. Confirm the appointment, offer Friday 4pm...', a model selector, and suggestion chips.

The first screen - describe the conversation you want to build

The first message is your onboarding. Type a one-line description of the agent’s job. Be concrete about the outcome and any fallbacks:
“Confirm an appointment - offer Friday 4pm, fall back to 5pm.” “Handle a refund request with a polite escalation path.”
The builder agent turns that into a first playbook: a persona, an opening line, and the checkpoints needed to reach the outcome.
Describe the goal and the guardrails, not the exact words. “Collect the order number before offering a refund” is a checkpoint the engine can enforce; scripting every sentence fights the model instead of steering it.

​
Start from a document or a URL

The composer has a + Attach button and accepts a pasted link:

Attach a document

Upload a PDF, TXT, MD, or DOCX (a returns policy, a product sheet, a script). The text is extracted and stored in the session, so the agent can re-read it on any later turn - not injected once and forgotten.

Paste a URL

Drop an https:// link in the composer. The page is fetched, extracted, and stored the same way. Great for “build an agent from our FAQ page.”
Sources persist for the whole session. The builder agent has list_sources and read_source tools, so you can say “check the returns doc again” three turns later and it will.

​
Step 2 - Pick the build model (optional)

The composer has a model selector. It defaults to a strong model that authors good playbooks out of the box, and each option is tagged for its strength - best for editing, most capable, top reasoning, fast:
Builder model dropdown listing Claude Sonnet 5 (best for editing), Claude Opus 4.8 (most capable), Claude Sonnet 4.6 (best for editing, selected), GPT-5.5 (top reasoning), Claude Haiku 4.5 (fast), and Gemini 3.5 Flash (beta).

The builder model picker - each model is labelled for the job it is best at

Override it only if you have a preference - it controls the builder, not the model your finished agent will run on (that is set in Test by voice).
Paste a website or doc link in the composer and the builder agent reads it before authoring - the hint under the box says so.

​
Step 3 - Watch it land in the Editor

Switch to the Editor tab to see the playbook the agent wrote. This is real, editable YAML:
Split view: left panel shows YAML editor with persona, env, journeys, and booking checkpoints including slots for name, service, and date; right panel shows Optimized by Agent with a refined version adding synonyms and advance conditions.

The Editor shows the YAML the builder agent authored - real, editable code you can tweak or let the agent refine

Two ways to refine from here:
1

Keep chatting

“Add a checkpoint that captures a callback number.” The agent edits the YAML and the Editor updates. This is the fastest loop.
2

Edit by hand

Type directly in the Editor. Validation runs as you type and flags a malformed checkpoint before you test it.

​
Step 4 - Save, import & export

The top bar holds the three file actions. The bar also shows a dirty/draft indicator, so you always know whether the current buffer is saved.
Saving keeps the playbook private to you. Making it visible to others, generating a shareable link, or deploying it happens later, in Publish & share - which offers Deploy as Voice Agent (a phone number) and Deploy as Endpoint (an API).

​
What “good” looks like before you test

  • A clear persona and an opening line a caller would find natural.
  • Checkpoints written as outcomes (done_when: the caller confirmed a time), not scripts.
  • The fallbacks you mentioned are present (offer 5pm; escalate politely).
  • Any slots you need to capture (order number, callback) have a checkpoint.

​
Next step

Test it by voice

Switch to the Preview tab, press to talk, and confirm the agent behaves like a real call before you harden or ship it.