mcpdirectory.dev
9,129 servers · updated 23 Sept

Dialfetch MCP server

community Active last commit 2 mo
{
  "mcpServers": {
    "dialfetch": {
      "type": "http",
      "url": "https://app.dialfetch.com/api/mcp"
    }
  }
}
file .mcp.json (in your project root)version 0.1.0

What it does

Dialfetch exposes phone calling capabilities to AI agents through the Model Context Protocol, allowing them to place real calls, navigate IVRs, and extract structured answers. It is for developers building AI agents that need to interact with phone-based systems to retrieve information like business hours, order status, or verification details.

  • Query business hours by calling a store or service and extracting closing time or availability information.
  • Verify licenses or insurance status by calling relevant organizations and extracting structured verification results.
  • Check order, claim, or account status by calling customer service lines and collecting structured answers about status.

Tools this server exposes

ToolDoes
place_callStart an asynchronous goal-directed phone call to a given number and return a call ID for tracking.
get_callRetrieve the status, outcome, extracted answers, transcript, and recording URL for a completed or in-progress call.
wait_for_callBlock until a call reaches a terminal state (completed, failed, or canceled) with optional timeout.
cancel_callCancel a queued or in-progress call by call ID.
list_callsList recent calls with optional filtering by status and pagination.
get_call_recordingRetrieve a short-lived presigned URL to the audio recording of a call.
list_phone_numbersList all caller-ID phone numbers owned by the account.
get_pricingRetrieve per-call and per-number pricing information.
get_credit_balanceRetrieve the current prepaid credit balance.

Before you enable it

An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.