Skip to content
AI Voice Agents
AI Voice Agents9 min read0 views

AI Voice Agent + Calendly Scheduling API: Book Meetings During the Call

Calendly's Scheduling API and Cal.com's open-source booking endpoints now let voice agents book meetings end-to-end without iframes. We walk through availability, booking, and webhook reconciliation.

Calendly shipped its Scheduling API and an MCP server in early 2026. Voice agents can finally book meetings without redirects, iframes, or "I will text you the link." Cal.com offers the same surface as open-source. The friction of "we'll find time later" is gone.

What the integration unlocks

flowchart LR
  Repo[GitHub repo] --> CI[GitHub Actions]
  CI --> Eval[Agent eval suite · PromptFoo]
  Eval -->|pass| Deploy[Deploy]
  Eval -->|fail| Block[Block PR]
  Deploy --> Prod[Production agent]
  Prod --> Trace[(LangSmith trace)]
  Trace --> Eval
CallSphere reference architecture

A discovery call ends. The CallSphere voice agent says "Tuesday 2pm or Thursday 10am with Priya?" The caller picks Thursday. Two seconds later the calendar invite is in their inbox, the Salesforce Opportunity has the meeting attached, the Slack #new-meetings channel has been notified, and the rep has a CRM-linked event on their day. Average booking-to-confirmation: 4-6 seconds. No follow-up email. No second touch.

How the API exposes it

Calendly's surface (paid plans only): GET /event_types to list bookable types, GET /event_type_available_times for slot lookup with start_time and end_time filters, and the new POST /scheduled_events (a.k.a. Create Event Invitee) to actually book. The Calendly MCP server exposes the same as MCP tools list_event_types, get_available_times, create_event. OAuth scope: default.

Cal.com's surface: GET /v2/event-types, GET /v2/slots/available, POST /v2/bookings. Self-hosted Cal.com lets you skip the OAuth dance entirely with API keys per organization.

Hear it before you finish reading

Talk to a live CallSphere AI voice agent in your browser — 60 seconds, no signup.

Try Live Demo →

Both support webhooks for invitee.created, invitee.canceled, invitee.no_show so your voice agent learns about reschedules without polling.

How CallSphere implements it

CallSphere's Healthcare vertical ships 14 tools including get_available_slots, schedule_appointment, and reschedule_appointment. The same primitives front Calendly, Cal.com, Athena, DrChrono, and ServiceTitan dispatch via a thin adapter — voice agents are unaware of which backend powers the schedule. Salon GlamBook integrates booking and payment in one turn. Real Estate OneRoof's Showing Coordinator agent books in-home visits via Cal.com when an agent's MLS calendar is hosted there. Sales product books discovery calls with Calendly directly from the browser dialer.

Pricing: $149 / $499 / $1499. 14-day trial. 22% affiliate.

Build steps

  1. For Calendly: register an OAuth application in Calendly Developer Portal. Implement the Authorization Code + PKCE flow.
  2. List the rep's Event Types once at config time and store the URIs.
  3. At call time, call event_type_available_times with a 7-day window. Cache for 60 seconds; users hate "that slot is gone" mid-conversation.
  4. Read 2-3 slot times back to the caller. Confirm one. Call POST /scheduled_events with the caller's name, email, phone, and selected start_time.
  5. Subscribe to the invitee.created webhook to confirm; if the booking fails (race), the voice agent re-prompts.
  6. Mirror the booking into your CRM as an Event with the Calendly URL attached.
  7. For Cal.com: same shape, simpler auth. Recommended for self-hosted multi-tenant CallSphere customers.

Code snippet

// Book a Calendly slot from the voice agent
const booking = await fetch("https://api.calendly.com/scheduled_events", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${calendlyOAuthToken}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    event_type: eventTypeUri,
    start_time: chosenSlot.iso,
    invitee: {
      name: caller.fullName,
      email: caller.email,
      phone_number: caller.e164,
      timezone: caller.timezone,
    },
    location: { kind: "phone_call", location: caller.e164 },
  }),
});

FAQ

Does the Calendly API require a paid plan? Yes, Scheduling API is gated to paid Calendly plans. Cal.com is free self-hosted or paid cloud.

Still reading? Stop comparing — try CallSphere live.

CallSphere ships complete AI voice agents per industry — 14 tools for healthcare, 10 agents for real estate, 4 specialists for salons. See how it actually handles a call before you book a demo.

How fresh is the slot data? Calendly recommends a 60-second TTL on cached availability. CallSphere uses 30s for high-traffic clinics.

What if the caller's slot collides with a same-second booking? Calendly returns a 409. The voice agent re-fetches and offers the next two slots in under one second.

Can we book group events? Yes — Calendly supports group event types. CallSphere's Healthcare vertical uses them for class-style appointments.

How do I demo it? Start a trial, connect Calendly or Cal.com in Settings, and run an outbound call to your own phone. See pricing.

Sources

Share

Try CallSphere AI Voice Agents

See how AI voice agents work for your industry. Live demo available -- no signup required.

Related Articles You May Like

AI Infrastructure

MCP Servers for SaaS Tools: A 2026 Registry Walkthrough for Voice Agent Teams

The public MCP registry crossed 9,400 servers in April 2026. Here is a curated walkthrough of the SaaS MCP servers CallSphere mounts in production, with OAuth 2.1 PKCE patterns.

Agentic AI

Building OpenAI Realtime Voice Agents with an Eval Pipeline (2026)

Build a working voice agent with the OpenAI Realtime API + Agents SDK, then bolt on an eval pipeline that catches barge-in failures, hallucinated grounding, and latency regressions.

Agentic AI

Voice Agent Quality Metrics in 2026: WER, Latency, Grounding, and the Ones Most Teams Miss

The full metric set for evaluating production voice agents — STT word error rate, end-to-end latency budgets, RAG grounding, prosody, and the metrics that actually correlate with retention.

Agentic AI

Online vs Offline Agent Evaluation: The Pre-Deploy / Post-Deploy Split

Offline evals catch regressions before deploy on a fixed dataset. Online evals catch real-world drift on live traffic. You need both — here is how we run them.

AI Strategy

Enterprise CIO Guide: Hippocratic AI — Healthcare Agents at Scale

Enterprise CIO Guide perspective on Hippocratic AI's deployment numbers show healthcare voice agents are moving from pilot to production across major US health systems.

AI Voice Agents

Vapi vs Retell vs Bland (2026): The Real Production Tradeoffs

Vapi (62M monthly calls), Retell (~600ms latency), Bland (volume scale). The honest 2026 comparison and where each is the wrong choice.