For the complete documentation index, see llms.txt.

How can AI agents schedule meetings?

AI agents can schedule meetings through Vennio using either MCP (Model Context Protocol) tools or the REST API directly. Vennio exposes 13 MCP tools covering availability lookup, booking creation, proposal negotiation, and network queries — purpose-built for agentic scheduling workflows.

The answer

Two integration paths for AI agents:

• MCP tools — connect Vennio to Claude, Cursor, or any MCP-compatible agent. The agent calls structured tools; Vennio handles calendar sync and booking confirmation automatically.
• REST API / SDK — your agent calls POST /v1/bookings directly. Works with any language or AI framework.

MCP tools available

Vennio's MCP server exposes 13 tools your agent can call:

• find_availability — bookable slots for a business
• find_mutual_availability — slots where multiple people are free
• create_booking — create a confirmed booking
• get_booking — retrieve booking details
• cancel_booking — cancel an existing booking
• propose_meeting — send time options to participants
• respond_to_proposal — accept, counter, or reject a proposal
• get_proposal — get proposal status and thread
• check_consent — verify calendar access permissions
• get_network — view network overview
• list_connections — list network connections
• enrich_meeting — add context to a meeting from web data
• find_mutual_connections — find shared network connections

Connect an agent via MCP

Add Vennio to your MCP client config. Here's the Claude Desktop configuration:

{
  "mcpServers": {
    "vennio": {
      "command": "npx",
      "args": ["-y", "@vennio/mcp"],
      "env": {
        "VENNIO_API_KEY": "vennio_sk_live_YOUR_KEY"
      }
    }
  }
}

Once connected, your agent can schedule meetings in natural language:

"Schedule a 30-minute call with customer@example.com tomorrow morning.
Check availability first, then create the booking and send confirmation."

Programmatic scheduling

For agents built with custom frameworks, call the REST API directly:

import { createVennioClient } from '@vennio/sdk'

const client = createVennioClient({ apiKey: process.env.VENNIO_API_KEY })

// Step 1: Find available slots
const availability = await client.availability.getSlots({
  query: {
    business_id: 'YOUR_BUSINESS_ID',
    duration_minutes: 30,
    timezone: 'America/New_York',
  },
})

const slots = availability.data.slots
if (slots.length === 0) {
  return 'No availability found in the requested window.'
}

// Step 2: Book the first suitable slot
const booking = await client.bookings.create({
  body: {
    business_id: 'YOUR_BUSINESS_ID',
    customer_email: 'customer@example.com',
    customer_name: 'Jane Doe',
    start_time: slots[0].start,
    end_time: slots[0].end,
    notes: 'Scheduled by AI assistant',
  },
})

return `Booked: ${booking.data.booking.id} at ${slots[0].start}`

Agent audit trail

When an agent creates a booking, pass agent_name in the request body (via the MCP create_booking tool or directly in the API). This appears in booking records so humans can see which agent took action.

curl -X POST https://api.vennio.app/v1/bookings \
  -H "Authorization: Bearer vennio_sk_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "business_id": "YOUR_BUSINESS_ID",
    "customer_email": "customer@example.com",
    "customer_name": "Jane Doe",
    "start_time": "2026-04-01T09:00:00Z",
    "end_time": "2026-04-01T09:30:00Z",
    "agent_name": "My Scheduling Agent v1"
  }'

Next steps

• AI Agent Integration (MCP) — full MCP setup guide with consent flow
• Firecrawl + Vennio MCP — enrich meetings with web research
• POST /v1/bookings — REST API reference
• GET /v1/availability/slots — availability reference