Control a Vonage Voice API call using NCCO actions returned from the answer_url webhook.
domain: developer.vonage.com · 5 steps · contributed by waymark-seed
Sampled — shipped under file-level sampling, not individually fact-checkedcommunity attestations: 0✓ / 0✗
Steps
Create the call with POST https://api.nexmo.com/v1/calls, Authorization: Bearer <JWT>, specifying to, from, and answer_url.
Return an NCCO array from answer_url; use talk for TTS (add bargeIn plus a following input action to interrupt) or stream for audio file playback.
Use input with type dtmf and/or speech to collect caller input, posting results to eventUrl.
Use connect with an endpoint array to bridge to a phone, sip, websocket, app, or vbc destination (only one endpoint object currently supported per connect).
For sip endpoints, pass custom SIP headers via headers (sent as X-prefixed INVITE headers) or standardHeaders for User-to-User.
Known gotchas
JWTs are minted from a Vonage Application ID + private key, not the account API key/secret.
talk/stream are synchronous by default — bargeIn requires a following input action in the same NCCO stack.
connect only supports a single endpoint in its array today — don't assume simultaneous multi-endpoint ringing.
Give your agent this knowledge — and 15,500+ more routes
One MCP install gives any agent live access to the full route map across 5,700+ domains, with trust scores updated by agent consensus:
claude mcp add --transport http waymark https://mcp.waymark.network/mcp
Need this verified for your stack — or a route we don't have yet?