Powsoo is a drop-in widget that adds visual intake cards to AI voice platforms. When a user is on a voice call with your agent, Powsoo renders synced cards in the browser so users can tap-to-confirm emails, phone numbers, dates, and other intake fields instead of spelling them out.

How do I add Powsoo to my voice platform?

Create a widget in the Powsoo dashboard and connect your Retell account, then drop one web component and one versioned script tag into your site. You add powsoo_show_card as a custom function on your agent. Setup takes about 5 minutes.

Which voice platforms does Powsoo support?

Retell today. Vapi and ElevenLabs support are coming soon. SMS fallback works for traditional PSTN calls.

Can I customize the look of the cards?

Yes. Powsoo supports full theming via JS configuration: colors, fonts, border radius, spacing. For deeper control, use Powsoo Atoms (React components) to build your own UI.

Where does the data go after a user confirms?

Powsoo fires webhooks the moment a card is confirmed, with the field name, value, and call session ID. Drop the payload straight into your CRM, database, or backend.

Does Powsoo work for phone calls?

Pure phone calls (PSTN) don't have a screen, so Powsoo's web widget can't render directly. Instead, Powsoo offers an SMS fallback: when the agent needs to confirm an email or phone number, the caller gets a one-tap SMS link.

Is Powsoo HIPAA / GDPR compliant?

Powsoo is built with healthcare and EU customers in mind. We're SOC 2 Type II in progress and offer DPAs and BAAs on the Pro plan and above.

What is a visual layer for AI voice agents?

Updated 2026-07-13

A visual layer is a browser widget that renders on-screen cards in sync with a voice agent, so the caller taps to confirm data such as emails, phone numbers, and dates instead of speaking them. It sits on top of platforms like Vapi, Retell, and ElevenLabs and turns a voice-only call into a voice-plus-screen (multimodal) one. The agent drives it: at the moment it needs a field confirmed, it calls a function with the value it heard, a card appears in the caller’s browser, and the confirmed result flows back into the same call session. This differs from a chat widget, which is a separate text thread the caller has to switch to. A visual layer stays tied to the live call and only appears for the fields where spoken capture is unreliable. When the caller taps Confirm, the clean value returns to your backend by webhook with the call session ID.

The problem it solves

Voice agents are good at conversation and poor at character strings. Speech-to-text turns an email or a phone number into a best guess, and one wrong character makes the value useless. A visual layer sidesteps transcription for those fields by moving them to the screen.

How it works

During a call, the agent decides it needs a field confirmed. It calls a function the visual layer provides. A card renders in the browser showing the value the agent heard. The user taps Confirm or edits it, and the confirmed value is sent back to your backend, tagged with the call session ID. The conversation never stops.

This is what makes it a multimodal interaction: voice and screen working on the same task at the same time, rather than a handoff between two separate tools.

Where Powsoo fits

Powsoo is a visual layer you add with one <powsoo-voice> web component and one script tag. It connects to Retell today, with Vapi and ElevenLabs coming soon: you add a powsoo_show_card function to your agent, and Powsoo fires a webhook the moment a card is confirmed. For calls with no screen, it falls back to an SMS link.

Frequently asked questions

How is a visual layer different from a chat widget?

A chat widget is a separate text conversation. A visual layer is tied to the live voice call: the agent triggers a card at the moment it needs a field confirmed, the card shows the value it heard, and the confirmed result flows back into the same call session.

Does a visual layer replace my voice platform?

No. It runs on top of Vapi, Retell, ElevenLabs, or any WebRTC voice platform. The platform keeps handling the conversation; the visual layer handles the fields where spoken capture is unreliable.

What can users confirm through a visual layer?

Anything error-prone to say aloud: email addresses, phone numbers, dates and dates of birth, reference codes, postcodes, and account numbers. The user reads and taps rather than spelling, which removes most transcription error.