Skip to main content
Glama

Place phone call

place_call
Destructive

Place one real outbound US phone call from a structured request. Returns call_id immediately; call track_call with it until the event is ended, then read get_call_transcript to judge whether the user's goal succeeded. outcome is the phone network, not task success. With delegation poll, the phone agent can ask you questions during the call, and you answer them through track_call. get_calling_guide topic outbound documents the request fields. The user may see a live card for the call.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "to": "<to>",
      "recipient": "<recipient>",
      "on_behalf_of": "<on_behalf_of>",
      "goal": "<goal>",
      "authority": "<authority>"
    },
    "name": "place_call"
  }
}

Input Schema

NameRequiredDescriptionDefault
toYesUS E.164 destination such as +14155550123. Look up official public business numbers; ask the user for private details or when listings conflict.
goalYesThe result wanted from this call. Include the full context needed to understand the outcome.
factsNoReference details the recipient may need for this goal, such as account or order numbers, dates, preferences, callback numbers, known menu paths, and prior-call outcomes. The phone agent knows only what the request contains.
voiceNoAudio voice: eric (default), jessica, sarah, or chris. Omission uses the saved voice.
handoffNoPresence enables loop-in. The assistant rings the account's loop-in contacts, or its verified phone when none are set. Include it on most calls. List every situation only the user can handle in handoff.when: one-time codes, account-holder-only policies, identity questions, payments, and decisions outside the authority. The assistant also loops in when the other side asks for the user or a real person.
languageNoLanguage to speak, when you know the recipient prefers one other than English. Speech recognition is tuned for it and English; the assistant can follow the other side into another listed language, less reliably. Omit for English. One of ar (Arabic), bn (Bengali), de (German), en (English), es (Spanish), fr (French), he (Hebrew), hi (Hindi), id (Indonesian), it (Italian), ja (Japanese), kn (Kannada), ko (Korean), mr (Marathi), ms (Malay), nl (Dutch), pl (Polish), pt (Portuguese), ta (Tamil), te (Telugu), th (Thai), tl (Tagalog), tr (Turkish), vi (Vietnamese), zh (Mandarin Chinese).
authorityYesCommitments the assistant may make and their limits. An empty allow list means ask and relay only.
questionsNoQuestions to ask the recipient, including alternatives and follow-up details. Anything the recipient must tell the user, such as a confirmation number or price, belongs here.
recipientYesBusiness, department, or person you expect to reach, including the location.
voicemailNoMailbox outcome: end (default) ends the call without leaving a message; leave_message carries only the words to convey, including authorized callback details. The agent leaves this message on the voicemail, so write it as the message itself, with exact names, numbers, and dates. Put no instructions to the agent inside message.
delegationNonone (default): the phone agent works from the request alone. poll: you stay with the call and answer what the phone agent needs to look up, by calling track_call until it returns ended. webhook: each question arrives at the webhook connected with connect_webhook as a call.delegation event with the call transcript so far, and you answer it with reply_on_call; it needs a connected webhook. Choose poll when you can stay with the whole call, or webhook when your host starts you from a webhook, and the call may need something you could not put in the request. Both still need complete facts.
on_behalf_ofYesPerson or organization the assistant represents. This identifies the requester; it grants no account access.
assistant_nameNoA plain assistant name, such as Jordan. Without it, the agent uses a default name, usually Alex.

Output Schema

NameRequiredDescriptionDefault
hintYes
call_idYes
messageYes
workflowYes
lifecycleYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.5/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The description adds substantial behavior beyond the annotations: a real outbound call is placed, call_id is returned immediately, outcome reflects the phone network rather than task success, delegation poll allows questions during the call, voicemail behavior is controlled elsewhere, and the user may see a live card. These details help an agent understand operational and side-effect implications in addition to the supplied destructive/open-world annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is front-loaded with the core action and return behavior, then proceeds through tracking, outcome interpretation, delegation, documentation, and user-visible effects. Each sentence contributes operational guidance without wasteful repetition.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a complex 13-parameter tool with nested objects and an output schema, the description covers the essential lifecycle, delegation choices, outcome caveat, voicemail implications, and user-visible card. The schema and output schema carry the field-level and return-value details, so no critical agent-facing context is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema already documents all 13 parameters in detail. The description adds workflow context around delegation and the distinction between phone-network outcome and task success, but it does not provide meaningful extra semantics for most request fields beyond pointing to get_calling_guide topic outbound.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a specific verb and resource: 'Place one real outbound US phone call from a structured request.' It distinguishes this tool from downstream tools by stating it returns call_id immediately and that track_call and get_call_transcript handle later stages.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description gives clear workflow context: call this tool, then poll track_call until ended, then inspect get_call_transcript. It also explains when to use delegation 'poll' versus 'webhook.' It does not explicitly compare this tool with sibling tools such as place_call_and_track or place_call_and_wait, so it falls short of a full when-to-use-alternatives treatment.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.