Ship the request.
Skip the staffing project.
Put the objective, context, and required fields in one API request. Let the call’s outcome flow straight back into your apps and agents.
Give your apps and agents a real human caller. Request a call through REST, MCP, or your favorite AI chat interface. A real person handles the conversation, and your apps and agents get the structured answers they need to move forward.
No hiring, no training, no managing staff. Just the power of a real human caller programmatically.
// POST /api/v1/call-requests · request body { "mode": "sandbox", "template": "appointment_confirmation", "company": { "name": "Northstar Home", "context": "Heating and cooling services. Only confirm the scheduled visit; do not change prices or make commitments." }, "recipient": { "name": "Maya Reynolds", "phone": "+12025550123" }, "appointment": { "service": "Annual heating and cooling tune-up", "date": "2026-10-15", "window": "10 AM–12 PM", "timezone": "America/New_York" } }
// Create a request through a sandbox MCP connection { "jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": { "name": "create_call_request", "arguments": { "idempotency_key": "visit-maya-2026-10-15-v1", "request": { "template": "appointment_confirmation", "company": { "name": "Northstar Home", "context": "Heating and cooling services. Only confirm the scheduled visit; do not change prices or make commitments." }, "recipient": { "name": "Maya Reynolds", "phone": "+12025550123" }, "appointment": { "service": "Annual heating and cooling tune-up", "date": "2026-10-15", "window": "10 AM–12 PM", "timezone": "America/New_York" } } } } }
// Completed sandbox result · response excerpt { "data": { "id": "req_example", "status": "completed", "result": { "mode": "sandbox", "simulated": true, "outcome": "confirmed", "answers": { "recording_consent": { "value": true, "evidence": "Yes, you may record this call." }, "appointment_confirmed": { "value": true, "evidence": "Yes, that time works." }, "appointment_status": { "value": "confirmed", "evidence": "Yes, that time works." }, "access_instructions": { "value": "Side entrance; gate code 2480.", "evidence": "Use the side entrance. The gate code is 2480." }, "pets_at_home": { "value": true, "evidence": "Yes, one dog. I will keep it in another room." } }, "recording": null } } }
Your application sends a REST API request.
For Northstar Home · heating & cooling services
Calling Maya Reynolds · October 15, 2026 · 10 AM–12 PM Eastern
The human caller is reviewing the request.
“Hello Maya Reynolds, I’m calling on behalf of Northstar Home to confirm your Annual heating and cooling tune-up. May we record this call?”
Animated preview of result.answers.*.value. The API includes evidence and returns answers after caller review.
Answers saved to your app database or CRM.
Structured results unlock the next step.
A follow-up, notification, or task starts.
Get the full recording, transcript, and results.
One clear brief. One atomized call request.
The result goes back to your apps & agents.
Consent requirements extend beyond sales. Under TCPA rules, AI-voice calls to mobile numbers generally require prior express consent—even for informational calls—unless an exemption applies. Covered telemarketing generally requires written consent.
AI-generated voices fall under TCPA artificial-voice rules. Violations can lead to $500 in statutory damages per call, potentially rising to $1,500 for willful or knowing violations.
A live, manually dialed call may be an option after the business verifies eligibility, Do Not Call restrictions, opt-outs, calling hours, and state laws.
Live calls still require an eligible contact. Political and charitable outreach can have additional rules.
Put the objective, context, and required fields in one API request. Let the call’s outcome flow straight back into your apps and agents.