AI Tools & MCP

# Core sending
send_email          — send a single transactional email
send_batch          — send up to 50 emails in one call
schedule_email      — schedule delivery at a future UTC timestamp
smart_send_at       — AI-optimised send-time per recipient

# Templates
create_template     — save a Handlebars template
generate_template   — generate HTML email from a plain-text brief

# Domains & infrastructure
verify_domain       — trigger DKIM/SPF verification for a domain
register_dedicated_ip — provision a warm dedicated IP
create_stream       — create a sending stream (transactional / marketing)

# Events & analytics
get_event_log       — fetch recent delivery events with filters
get_analytics       — open / click / bounce rates for a date range

# AI utilities
summarize           — summarise an email thread or body
ask_ai              — ask an open question about your email data
auto_label          — classify & label an incoming message

# Contacts & team
list_contacts       — list / search contacts with pagination
invite_user         — invite a team member to your workspace

# Files
upload_file         — upload an attachment (base64 or URL)

# Utilities
ping                — health-check the MCP bridge

{
  "mcpServers": {
    "mailgrid": {
      "command": "npx",
      "args": ["-y", "@mailgrid/mcp"],
      "env": {
        "MAILGRID_API_KEY": "mb_live_..."
      }
    }
  }
}

{
  "mcpServers": {
    "mailgrid": {
      "command": "npx",
      "args": ["-y", "@mailgrid/mcp"],
      "env": {
        "MAILGRID_API_KEY": "mb_live_..."
      }
    }
  }
}

{
  "mcpServers": {
    "mailgrid": {
      "command": "npx",
      "args": ["-y", "@mailgrid/mcp"],
      "env": {
        "MAILGRID_API_KEY": "mb_live_..."
      }
    }
  }
}

pip install mailgrid-langchain

from langchain.agents import create_react_agent, AgentExecutor
from langchain_anthropic import ChatAnthropic
from mailgrid_langchain import MailgridToolkit

toolkit = MailgridToolkit(
    api_key="mb_live_...",
    from_email="you@yourdomain.com",
)

llm   = ChatAnthropic(model="claude-sonnet-4-6")
tools = toolkit.get_tools()
agent = create_react_agent(llm, tools)
executor = AgentExecutor(agent=agent, tools=tools, verbose=True)

result = executor.invoke({
    "input": "Send a welcome email to alice@example.com and report the open rate for last week."
})
print(result["output"])

pip install mailgrid-llamaindex

from llama_index.core.agent import ReActAgent
from llama_index.llms.anthropic import Anthropic
from mailgrid_llamaindex import MailgridAgentTools

tools = MailgridAgentTools(
    api_key="mb_live_...",
    from_email="you@yourdomain.com",
).as_tools()

llm   = Anthropic(model="claude-sonnet-4-6")
agent = ReActAgent.from_tools(tools, llm=llm, verbose=True)

response = agent.chat(
    "How many contacts opened emails in the last 30 days? "
    "Send a re-engagement campaign to those who haven't opened in 14 days."
)
print(response)

npm install @mailgrid/openai-functions

import OpenAI from "openai";
import { functions, executeFunction } from "@mailgrid/openai-functions";

const client = new OpenAI();

const messages: OpenAI.Chat.ChatCompletionMessageParam[] = [
  {
    role: "system",
    content: "You are an email assistant. Use the available tools to help the user.",
  },
  {
    role: "user",
    content: "Send a confirmation email to bob@example.com and show me last week's stats.",
  },
];

while (true) {
  const response = await client.chat.completions.create({
    model: "gpt-4o",
    messages,
    tools: functions.map((f) => ({ type: "function", function: f })),
    tool_choice: "auto",
  });

  const choice = response.choices[0];
  messages.push(choice.message);

  if (choice.finish_reason === "stop") {
    console.log(choice.message.content);
    break;
  }

  for (const call of choice.message.tool_calls ?? []) {
    const result = await executeFunction(
      call.function.name,
      JSON.parse(call.function.arguments),
      { apiKey: process.env.MAILGRID_API_KEY! }
    );
    messages.push({
      role: "tool",
      tool_call_id: call.id,
      content: JSON.stringify(result),
    });
  }
}