Skip to main content
Glama

Hail Communication for AI

Create an agent

create_agent

Save an agent for calls, texts, and email replies.

A handover contact's phone number is hidden from the call agent. Saving an agent does not change any phone number's routing. Returns the saved agent record or an error.

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": {
      "name": "<name>",
      "system_prompt": "<system_prompt>"
    },
    "name": "create_agent"
  }
}

Input Schema

NameRequiredDescriptionDefault
nameYesAgent display name.
toolsNoNames of the tools the agent may use during a call or reply. Omitted: every tool the workspace's channels support. An empty list: none.
languageNoLanguage the agent speaks, as an ISO 639-1 code such as ``es``. Omitted: English.
voice_idNoText-to-speech voice ID from the voice provider that serves the call, such as an ElevenLabs voice ID. Omitted: the workspace's default voice.
llm_modelNoModel ID of the calling agent. Optional. Recorded for analytics only.
http_toolsNoUp to ten customer APIs the agent may call during calls, texts and email. Each entry: ``name`` (letters, digits, _), ``description`` (read by the model), ``url`` (public https), optional ``parameters`` (JSON Schema, type object), ``headers`` (static, stored encrypted, returned as names only), ``timeout_seconds`` (1 to 10, default 5) and ``wait_message`` (said on a call while waiting). The list replaces the saved one; an empty list removes it. On update, an entry without ``headers`` keeps the saved headers of the same name while the url keeps its host; a url on a new host needs ``headers`` again. The first save returns ``http_tool_secret`` once; requests carry ``X-Hail-Signature`` signed with it.
sms_enabledNoWhether this agent answers texts on numbers whose texts route to it.
ai_disclosureNoSpeak the AI disclosure line first on every call.
email_enabledNoWhether this agent replies to mail on addresses routed to it.
first_messageNoGreeting spoken after the AI disclosure. Without one, an inbound agent waits for the caller.
system_promptYesInstructions that define the agent's behavior on calls, texts and email replies.
voice_enabledNoWhether this agent answers calls on numbers whose calls route to it.
handover_contactsNoUp to ten ``{contact_id, note}`` entries for people a call can be handed over to. ``contact_id`` is an ID from ``list_contacts``, ``lookup_contact`` or ``create_contact`` (a UUID or ``member:<user_id>``); the contact needs a phone number. ``note`` (1 to 200 characters) says when the agent hands over to that person, such as ``billing questions``.
recording_enabledNoRecording for this agent's calls. ``true`` or ``false`` overrides the workspace setting.
ai_disclosure_lineNoCustom AI disclosure line replacing the workspace line. ``{org}`` expands to the organization name.
max_duration_secondsNoLongest a call with this agent may last, in seconds, counted from pickup. At the limit the agent says goodbye and hangs up. Omitted: the workspace's call limit.
handover_max_duration_secondsNoLongest a call may last after a handover contact answers, in seconds, counted from that answer. Omitted: 1800 (30 minutes).

Output Schema

NameRequiredDescriptionDefault

No arguments

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • addedInput schema / properties / http_tools
      Added value: +{
      +  "anyOf": [
      +    {
      +      "items": {
      +        "additionalProperties": true,
      +        "type": "object"
      +      },
      +      "type": "array"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "description": "Up to ten customer APIs the agent may call during calls, texts and email. Each entry: ``name`` (letters, digits, _), ``description`` (read by the model), ``url`` (public https), optional ``parameters`` (JSON Schema, type object), ``headers`` (static, stored encrypted, returned as names only), ``timeout_seconds`` (1 to 10, default 5) and ``wait_message`` (said on a call while waiting). The list replaces the saved one; an empty list removes it. On update, an entry without ``headers`` keeps the saved headers of the same name while the url keeps its host; a url on a new host needs ``headers`` again. The first save returns ``http_tool_secret`` once; requests carry ``X-Hail-Signature`` signed with it.",
      +  "title": "Http Tools"
      +}
  2. Changed1 schema field changed
    • addedInput schema / properties / llm_model
      Added value: +{
      +  "description": "Model ID of the calling agent. Optional. Recorded for analytics only.",
      +  "type": "string"
      +}
  3. Changed11 schema fields changed
    • changedInput schema / properties / email_enabled / description
      Previous value: -"Answer email on addresses that route email to this agent."New value: +"Whether this agent replies to mail on addresses routed to it."
    • changedInput schema / properties / handover_max_duration_seconds / anyOf
      Previous value: -[
      -  {
      -    "type": "integer"
      -  },
      -  {
      -    "type": "null"
      -  }
      -]New value: +[
      +  {
      +    "maximum": 3600,
      +    "minimum": 60,
      +    "type": "integer"
      +  },
      +  {
      +    "type": "null"
      +  }
      +]
    • changedInput schema / properties / handover_max_duration_seconds / description
      Previous value: -"Length limit in seconds for a call handed over to a person, 60 to 3600. The default is 1800."New value: +"Longest a call may last after a handover contact answers, in seconds, counted from that answer. Omitted: 1800 (30 minutes)."
    • changedInput schema / properties / language / anyOf
      Previous value: -[
      -  {
      -    "type": "string"
      -  },
      -  {
      -    "type": "null"
      -  }
      -]New value: +[
      +  {
      +    "enum": [
      +      "ar",
      +      "bg",
      +      "bn",
      +      "cs",
      +      "da",
      +      "de",
      +      "el",
      +      "en",
      +      "es",
      +      "fi",
      +      "fr",
      +      "gu",
      +      "he",
      +      "hi",
      +      "hr",
      +      "hu",
      +      "id",
      +      "it",
      +      "ja",
      +      "kn",
      +      "ko",
      +      "mr",
      +      "ms",
      +      "nl",
      +      "no",
      +      "pl",
      +      "pt",
      +      "ro",
      +      "ru",
      +      "sk",
      +      "sv",
      +      "ta",
      +      "te",
      +      "th",
      +      "tl",
      +      "tr",
      +      "uk",
      +      "vi",
      +      "zh"
      +    ],
      +    "type": "string"
      +  },
      +  {
      +    "type": "null"
      +  }
      +]
    • changedInput schema / properties / language / description
      Previous value: -"Spoken language as a lowercase ISO 639-1 code, such as ``es``."New value: +"Language the agent speaks, as an ISO 639-1 code such as ``es``. Omitted: English."
    • changedInput schema / properties / max_duration_seconds / anyOf
      Previous value: -[
      -  {
      -    "type": "integer"
      -  },
      -  {
      -    "type": "null"
      -  }
      -]New value: +[
      +  {
      +    "maximum": 3600,
      +    "minimum": 60,
      +    "type": "integer"
      +  },
      +  {
      +    "type": "null"
      +  }
      +]
    • changedInput schema / properties / max_duration_seconds / description
      Previous value: -"Call length limit in seconds, 60 to 3600."New value: +"Longest a call with this agent may last, in seconds, counted from pickup. At the limit the agent says goodbye and hangs up. Omitted: the workspace's call limit."
    • changedInput schema / properties / sms_enabled / description
      Previous value: -"Answer texts on numbers that route texts to this agent."New value: +"Whether this agent answers texts on numbers whose texts route to it."
    • changedInput schema / properties / tools / description
      Previous value: -"Names of the tools the agent may use, checked against the server's tool registry. An empty list disables tools."New value: +"Names of the tools the agent may use during a call or reply. Omitted: every tool the workspace's channels support. An empty list: none."
    • changedInput schema / properties / voice_enabled / description
      Previous value: -"Answer calls on numbers that route calls to this agent."New value: +"Whether this agent answers calls on numbers whose calls route to it."
    • changedInput schema / properties / voice_id / description
      Previous value: -"Text-to-speech voice ID, applied by whichever voice provider serves the call."New value: +"Text-to-speech voice ID from the voice provider that serves the call, such as an ElevenLabs voice ID. Omitted: the workspace's default voice."
  4. First observed

TDQS

B3.4/5.0
Behavior3/5

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

Annotations already declare this is a non-readonly, non-destructive, non-idempotent write, so the safety profile is covered. The description adds two genuinely useful facts beyond that: saving does not alter phone-number routing, and a handover contact's number is hidden from the call agent. It omits other side effects the parameters imply, such as the one-time http_tool_secret being surfaced only on the first save.

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

Conciseness4/5

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

Three short, front-loaded sentences with no filler; the channel scope comes first and the two caveats follow. The only weakness is that "Save" is slightly at odds with the name's "create" framing and could mislead about upsert behavior.

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 17-parameter creation tool, the structured data carries the weight: full schema coverage, an output schema for the returned record, and annotations for the safety profile. The description fills the remaining gap an agent needs — side effects on routing and the return shape — so little 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 every one of the 17 parameters is documented in the schema itself, including defaults and the http_tools/headers persistence rules. The description adds no parameter-level meaning beyond that, which makes the baseline 3 appropriate.

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

Purpose4/5

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

"Save an agent for calls, texts, and email replies" names a concrete resource and the channel scope the agent covers, which is clearer than the bare name. It does not distinguish itself from update_agent or get_agent, so an agent must rely on the name to know this is the creation path rather than the modification path.

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

Usage Guidelines2/5

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

There is no when-to-use guidance and no mention of the obvious alternative, update_agent, for modifying an existing agent. Nothing states prerequisites either, such as which workspace settings or permissions must already exist before saving.

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.

Resources