Skip to main content
Glama
ghl-mcp-server-v2

ghl-mcp-server-v2

by zackscriven

ghl_conversation_send_message

Sends an SMS, email, WhatsApp, or other message to a contact in an existing GoHighLevel conversation thread.

Instructions

PHI-adjacent: sends a message on a real conversation thread. DELIBERATE OVERRIDE of the broken upstream spec (2026-08-26): the official conversations-v3.json declares subType as an empty object schema (the live API validates it as a string enum, so {} 422s with 'subType must be a valid enum value') and marks [type, subType, contactId, status] as required (forcing subType/status is what makes every call fail — a plain SMS needs only type + contactId + message). This tool therefore requires only type and contactId; subType and status are optional strings and are OMITTED from the outbound request entirely when not supplied (the API validates the key if present). The spec enumerates no valid subType values — pass one only if you know it from GHL docs. Populate type-appropriate fields alongside: message (+ fromNumber/toNumber) for SMS-style channels; html/subject/emailTo/emailCc for Email; attachments as URLs. If the spec extractor is re-run, keep this patchBody override — do not let regeneration silently restore the broken schema. Send a new message Post the necessary fields for the API to send a new message. Endpoint: POST /conversations/messages (Version header: v3; source: v3/conversations-v3.json) OAuth scopes: conversations/message.write

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": {
      "body": "<body>"
    },
    "name": "ghl_conversation_send_message"
  }
}

Input Schema

NameRequiredDescriptionDefault
bodyYesRequest body (schema carried verbatim from the official OpenAPI spec).

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed16 schema fields changedv2.1.0
    • removedInput schema / properties / body / properties / customSubtypeId
      Removed value: -{
      -  "description": "Custom subtype ID for email unsubscription preferences. Only applies to email messages.",
      -  "example": "507f1f77bcf86cd799439011",
      -  "type": "string"
      -}
    • removedInput schema / properties / body / properties / forward
      Removed value: -{
      -  "allOf": [
      -    {
      -      "properties": {
      -        "emailMessageId": {
      -          "description": "Email Message ID of the specific email being forwarded (source) - Required for single email forward, ignored for thread forward",
      -          "example": "rnGyqh2F6uBrIkfhFo9A",
      -          "type": "string"
      -        },
      -        "forwardWholeThread": {
      -          "description": "Specify if forwarding the whole thread or just a single email",
      -          "example": false,
      -          "type": "boolean"
      -        },
      -        "isForwarded": {
      -          "description": "Specify if this is a forwarded email",
      -          "example": true,
      -          "type": "boolean"
      -        },
      -        "messageId": {
      -          "description": "Message ID of the email thread being forwarded (source) - REQUIRED for forwarding",
      -          "example": "t22c6DQcTDf3MjRhwf77",
      -          "type": "string"
      -        },
      -        "recipientContactId": {
      -          "description": "Contact ID of recipient when forwarding (destination)",
      -          "example": "DEF56h2F6uBrIkfXYacd",
      -          "type": "string"
      -        },
      -        "recipientConversationId": {
      -          "description": "Conversation ID of recipient when forwarding (destination)",
      -          "example": "GHI78h2F6uBrIkfXYefg",
      -          "type": "string"
      -        },
      -        "sourceContactId": {
      -          "description": "Contact ID where the forwarded email originated from (source) - Auto-populated if not provided",
      -          "example": "ABC12h2F6uBrIkfXYazb",
      -          "type": "string"
      -        },
      -        "sourceConversationId": {
      -          "description": "Conversation ID where the forwarded email originated from (source) - Auto-populated if not provided",
      -          "example": "XYZ12h2F6uBrIkfXYacd",
      -          "type": "string"
      -        },
      -        "toEmail": {
      -          "description": "Email address to forward to (destination)",
      -          "example": "forward@example.com",
      -          "type": "string"
      -        }
      -      },
      -      "required": [
      -        "isForwarded"
      -      ],
      -      "type": "object"
      -    }
      -  ],
      -  "description": "Forwarding configuration for emails",
      -  "example": {
      -    "emailMessageId": "rnGyqh2F6uBrIkfhFo9A",
      -    "forwardWholeThread": false,
      -    "isForwarded": true,
      -    "messageId": "t22c6DQcTDf3MjRhwf77",
      -    "recipientContactId": "DEF56h2F6uBrIkfXYacd",
      -    "toEmail": "forward@example.com"
      -  }
      -}
    • addedInput schema / properties / body / properties / mentions
      Added value: +{
      +  "description": "Array of user IDs mentioned in the message. Required for InternalComment type. User IDs correspond to team members tagged with `@username<userId>id</userId>` format in the message text.",
      +  "example": [
      +    "userId123",
      +    "userId456"
      +  ],
      +  "items": {
      +    "type": "string"
      +  },
      +  "type": "array"
      +}
    • changedInput schema / properties / body / properties / message / description
      Previous value: -"Text content of the message"New value: +"Text content of the message. For InternalComment type, use `@username<userId>actualUserId</userId>` format to mention team members. The mentioned user IDs must also be included in the mentions array."
    • removedInput schema / properties / body / properties / optimizationPeriod
      Removed value: -{
      -  "description": "Optimization period in hours (24h, 48h, or 72h)",
      -  "enum": [
      -    "24h",
      -    "48h",
      -    "72h"
      -  ],
      -  "example": "24h",
      -  "type": "string"
      -}
    • changedInput schema / properties / body / properties / status / description
      Previous value: -"Message status"New value: +"Message status. OPTIONAL — reads like a response/inbound-recording field; do not set on a normal outbound send."
    • removedInput schema / properties / body / properties / status / example
      Removed value: -"delivered"
    • changedInput schema / properties / body / properties / subType / description
      Previous value: -"Type of message being sent"New value: +"Message sub-type. OPTIONAL — omit for normal sends. The official spec declares this as an empty object schema and enumerates no valid values (spec gap); the live API validates it as a string enum when the key is present, so only pass a value you know to be valid."
    • removedInput schema / properties / body / properties / subType / example
      Removed value: -"Email"
    • changedInput schema / properties / body / properties / subType / type
      Previous value: -"object"New value: +"string"
    • changedInput schema / properties / body / properties / type / enum
      Previous value: -[
      -  "SMS",
      -  "RCS",
      -  "Email",
      -  "WhatsApp",
      -  "IG",
      -  "FB",
      -  "Custom",
      -  "Live_Chat",
      -  "TIKTOK"
      -]New value: +[
      +  "SMS",
      +  "Email",
      +  "WhatsApp",
      +  "IG",
      +  "FB",
      +  "Custom",
      +  "Live_Chat",
      +  "InternalComment"
      +]
    • addedInput schema / properties / body / properties / userId
      Added value: +{
      +  "description": "Use this field to specify the user who is making the internal comment when type is 'InternalComment'. If not provided, the comment will be attributed to the system or default user.",
      +  "example": "user123",
      +  "type": "string"
      +}
    • removedInput schema / properties / body / properties / usesNativeSchedulingAi
      Removed value: -{
      -  "description": "Whether the scheduled email uses native AI for the email scheduling ",
      -  "example": false,
      -  "type": "boolean"
      -}
    • addedInput schema / properties / body / properties / whatsapp
      Added value: +{
      +  "allOf": [
      +    {
      +      "properties": {
      +        "media": {
      +          "allOf": [
      +            {
      +              "properties": {
      +                "name": {
      +                  "description": "File name shown to the recipient.",
      +                  "example": "sample.png",
      +                  "type": "string"
      +                },
      +                "type": {
      +                  "description": "Kind of media being sent.",
      +                  "enum": [
      +                    "image",
      +                    "video",
      +                    "audio",
      +                    "document"
      +                  ],
      +                  "example": "image",
      +                  "type": "string"
      +                },
      +                "url": {
      +                  "description": "Publicly reachable URL of the file. Upload via POST /conversations/messages/upload and use the returned URL.",
      +                  "example": "https://static-assets.internal.usercontent.site/conversations-assets/location/<locationId>/conversations/contact/<contactId>/<uuid>.png",
      +                  "type": "string"
      +                }
      +              },
      +              "required": [
      +                "type",
      +                "url",
      +                "name"
      +              ],
      +              "title": "WhatsAppAttachment",
      +              "type": "object"
      +            }
      +          ],
      +          "description": "Media to send. Required when `type` is `media`."
      +        },
      +        "type": {
      +          "description": "Send `media` to deliver an image, video, audio or document. Omit the `whatsapp` object entirely for a text-only message.",
      +          "enum": [
      +            "media"
      +          ],
      +          "example": "media",
      +          "type": "string"
      +        }
      +      },
      +      "required": [
      +        "type"
      +      ],
      +      "title": "WhatsappPayloadDto",
      +      "type": "object"
      +    }
      +  ],
      +  "description": "WhatsApp specific payload. Use this to send media (image, video, audio, document) on a WhatsApp message. Only applies when `type` is `WhatsApp`.",
      +  "example": {
      +    "media": {
      +      "name": "sample.png",
      +      "type": "image",
      +      "url": "https://static-assets.internal.usercontent.site/conversations-assets/location/<locationId>/conversations/contact/<contactId>/<uuid>.png"
      +    },
      +    "type": "media"
      +  }
      +}
    • changedInput schema / properties / body / required
      Previous value: -[
      -  "type",
      -  "subType",
      -  "contactId",
      -  "status"
      -]New value: +[
      +  "type",
      +  "contactId"
      +]
    • addedInput schema / properties / body / title
      Added value: +"SendMessageBodyDto"
  2. First observedv2.0.0

TDQS

A3.9/5.0
Behavior4/5

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

Annotations already cover the mutation/safety profile (readOnlyHint=false, idempotentHint=false, openWorldHint=true), so the bar is lower; the description still adds real context: PHI-adjacent data, a live external delivery, the required OAuth scope, and the important detail that subType/status are omitted from the request entirely when absent because the API validates the key if present.

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

Conciseness3/5

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

The key point (PHI-adjacent real send) is front-loaded, but a large share of the text is maintainer-facing meta-commentary — the 2026-08-26 override note, the spec-extractor regeneration warning, the empty-object-schema forensics — which does not help an agent decide or call correctly and dilutes the operational guidance.

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

Completeness4/5

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

For a mutation tool with a rich nested schema and no output schema, the description covers endpoint, version header, scopes, channel field mapping, and the subType/status quirk. It omits what comes back (message id, status) and any rate-limit or delivery-failure behavior, which an agent calling a live messaging endpoint would benefit from knowing.

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 100%, so the baseline is 3; the description earns above that by explaining which fields belong together per channel type and by clarifying the effective required set (type + contactId only) versus the upstream spec's inflated required list. It does not document the whatsapp media sub-object or threadId/reply fields beyond what the schema already says.

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?

The description states a concrete verb and resource — sends a message on a real conversation thread — and anchors it to a specific endpoint (POST /conversations/messages, v3). It does not differentiate from close siblings such as ghl_conversation_add_outbound_message or ghl_conversation_add_inbound_message, which an agent could plausibly confuse it with.

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 concrete per-channel guidance (message/fromNumber/toNumber for SMS-style, html/subject/emailTo/emailCc for Email, attachments as URLs) and explicitly says to pass subType only if you know a valid value. It stops short of naming when to prefer a sibling tool or when not to send at all.

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

Deploy Server

Other Tools