# AgentPhone: phone identity and messaging infrastructure for AI agents

Canonical: https://mudpie.ai/companies/agentphone/
Breadcrumb: [Home](https://mudpie.ai/) / [Companies](https://mudpie.ai/companies/) / [AgentPhone: phone identity and messaging infrastructure for AI agents](https://mudpie.ai/companies/agentphone/)
Author: Ali Abouelatta (https://mudpie.ai/authors/ali-abouelatta/)
Published: 2026-09-19
Updated: 2026-09-19
Research type: Company profile
Method: Company and accelerator sources checked 2026-09-19. Product claims are attributed to their sources; this is research, not a hands-on product trial.

AgentPhone gives an agent a phone identity instead of asking every builder to rebuild telecom.

## What it does

AgentPhone describes itself as a phone-number API for AI agents. The [current homepage](https://agentphone.ai/) shows agents provisioning numbers, sending messages, receiving calls, using real-time transcription, and connecting through REST, SDKs, or MCP.

The buyer is a team building an agent that needs to reach people outside the browser: sales follow-up, support, 2FA, scheduling, delivery, or agent-to-agent coordination.

| Fact | What the public source says |
| --- | --- |
| Buyer | AI-agent and product teams |
| Product | Phone numbers, SMS, calls, webhooks, transcripts, MCP, SDKs |
| Coverage | The homepage says US and Canadian numbers are available, with more countries planned |
| Founder context | YC says Meet Modi worked on AI-agent infrastructure at Meta and Manav Modi worked on the Vogue app |

## Why it is useful

The product boundary is clear. AgentPhone handles the phone number, messaging and call events, threading, transcription, and a unified webhook. The builder can focus on the agent instead of stitching together telecom providers and separate voice and SMS infrastructure.

The MCP support is also practical. The docs describe using AgentPhone from MCP-compatible clients, while the API supports Python, Node, REST, and direct tool flows. That makes the buyer a developer who wants the same agent to act over web and phone channels.

The founder context fits the wedge. YC describes Meet Modi as having built AI-agent infrastructure on WhatsApp at Meta, while Manav Modi worked on the Vogue app. The backgrounds explain the connectivity focus, not the reliability of every telecom path.

## What I would resolve

Phone infrastructure brings obligations that a demo can hide. A buyer should resolve geography, number provisioning, carrier behavior, consent, opt-out, rate limits, recording, transcript retention, and handoff. The public docs say usage is pay-as-you-go, but the exact current rate card deserves a live check.

The product is a fit for an agent that genuinely needs to call or text. If the workflow stays inside email or a web app, AgentPhone adds a communications surface the team may not need.

Short version: shortlist AgentPhone when phone identity is the missing capability in an agent product. The API is the easy part. The hard part is operating communications responsibly at scale.

## Sources checked — 2026-09-19

- [YC profile](https://www.ycombinator.com/companies/agentphone)
- [AgentPhone homepage](https://agentphone.ai/)
- [AgentPhone docs](https://docs.agentphone.ai/welcome)
- [AgentPhone pricing](https://agentphone.ai/pricing)
- [YC launch](https://www.ycombinator.com/launches/QNE-agentphone-phone-numbers-for-ai-agents)

## Cohort context

AgentPhone is listed in Spring 2026. In our 2026-09-18 directory snapshot, 112 of 193 listed companies in that cohort have YC’s primary industry label B2B (58.0%). This is a current-directory comparison, not an original intake count or a performance ranking. [Nine-cohort dataset](https://mudpie.ai/research/yc-cohorts-2026-09-19.json).

## Public website snapshot

Observed 2026-09-19T16:16:02.378Z in raw homepage HTML. This records visible metadata and advertised links, not agent execution or product quality.

| Signal | Homepage observation |
| --- | --- |
| Product description metadata | Observed |
| Canonical link | Observed |
| H1 or H2 heading | Observed |
| Typed structured data | Observed |
| Docs/developer link | Observed |
| Pricing link | Observed |
| llms.txt link | Not observed in this response |
| Markdown alternate | Not observed in this response |

[Public observations](https://mudpie.ai/research/yc-homepage-links-2026-09-19.json) · [Collection method](https://mudpie.ai/research/yc-homepage-methods/README.md). Missing links here do not establish that a capability or file is absent elsewhere.


## Author disclosure

I cofound Lazyweb and publish Mudpie. This is an owner-written publication, not an independent testing organization. Research notes distinguish observations, sourced reporting and editorial judgment.