Tool reference

Four tools for the call.

Updated August 15, 2026

DialMCP exposes four MCP tools. Calls are asynchronous by design: place_call returns a call ID immediately, then your agent uses get_call to follow progress and retrieve the final result.

`place_call`

Start an outbound call to a US or Canadian number with a stated objective. The request includes the destination and what the agent should try to accomplish. The tool returns a call ID while the call continues in the background.

`get_call`

Fetch the current state of a call by ID. While the call is running, the response includes status and live progress. When it finishes, the response includes:

A structured resolution: achieved, partially_achieved, or not_achieved.
A summary, commitments made by the other party, and suggested follow-ups.
A turn-by-turn transcript.
A signed link to the audio recording.

`end_call`

Hang up an in-progress call immediately. Use it when the objective is complete or the conversation should stop.

`list_calls`

List recent calls for the authenticated user with their statuses and outcomes. Use it to find prior call IDs or review recent activity.

Typical lifecycle

Call place_call with a clear, allowed objective.
Poll get_call until the call reaches a terminal state.
Read the structured resolution and transcript, then report the result to the user.
Call end_call if the user wants to stop before completion.

Guardrails apply to every tool: calls disclose that they are AI-assisted and recorded, use the verified caller ID, respect destination-local hours and server-side rate limits, and cannot reach blocked ranges. Review Safety for the full model.

Need to connect first? Follow the installation guide. Need to understand sign-in? Read OAuth and phone verification. Want the end-to-end walkthrough? See how to make a phone call with an AI assistant. Comparing a dedicated app to an agent in your existing client? See AI phone call app vs. agent. Checking launch pricing? See free during launch.