Skip to main content
Glama

Reply on a live call

reply_on_call
DestructiveIdempotent

Answer a call.delegation event while its call is live. The phone agent passes your text on in its own words, usually in a sentence or two, and keeps the rest to share if asked. If the answer will take more than about 20 seconds, first send a short progress reply with final false, then the answer; after 45 seconds with nothing from you, the phone agent carries on without you. Another reply to an answered delegation adds to the answer, and an exact copy of an earlier reply is not passed on again. delivered false means the call has ended or the delegation is unknown, so tell the user directly instead.

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": {
      "call_id": "<call_id>",
      "delegation_id": "<delegation_id>",
      "text": "<text>"
    },
    "name": "reply_on_call"
  }
}

Input Schema

NameRequiredDescriptionDefault
textYesYour answer, with the answer first and names and numbers exact, in plain words without markdown or links. Up to 2,000 characters.
finalNofalse for a short progress reply that keeps the delegation open, such as that you are still checking. Default true.
call_idYesdata.call.id from the call.delegation event.
delegation_idYesdata.delegation.id from the call.delegation event.

Output Schema

NameRequiredDescriptionDefault
deliveredYesTrue when the reply was recorded on the live call. A reply the phone agent never passes on, such as one sent after the user was connected, appears in call.completed data.delegations.unspoken_replies.
reply_resultYesaccepted, call_ended when the call is over, or unknown_request when the call has no such delegation.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.5/5.0
Behavior4/5

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

Beyond the annotations, it explains that the phone agent paraphrases the text, keeps extra detail for follow-up, times out after about 45 seconds, and treats duplicate replies specially. The destructiveHint annotation is not explained, leaving one behavioral trait opaque, but the description is otherwise unusually transparent.

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, then covers timing, duplicate handling, and failure handling in a few dense but purposeful sentences. No sentence is redundant.

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 live-call reply tool with an output schema, it covers the interaction model, timing expectations, progress replies, duplicate behavior, and the key failure case. An agent has enough context to call it correctly without needing to infer critical behavior.

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

Parameters4/5

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

Schema coverage is already 100%, so the schema carries the baseline parameter definitions. The description still adds meaning, especially for final=false as a progress reply that keeps the delegation open and for the failure meaning of delivered=false.

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 states a specific verb and resource: 'Answer a call.delegation event while its call is live.' It is immediately distinguishable from call-management siblings such as place_call, hangup_call, and get_call_transcript.

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?

It gives clear operational guidance for live delegations, including when to send a progress reply and what to do when delivered is false because the call has ended. It does not explicitly name alternative tools, but no sibling appears to serve the same reply role.

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.