Email-Enabled Chatflows with Flowise

Give your Flowise chatbots and agent flows the ability to send, read, and manage email through MultiMail — with human oversight ensuring safe delivery.


Flowise is an open-source low-code tool for building LLM orchestration flows and AI agents. Its drag-and-drop UI enables creating chatbots, RAG pipelines, and multi-agent systems with built-in API endpoints. MultiMail provides the email infrastructure that extends Flowise chatflows beyond chat into real email communication.

By connecting MultiMail to Flowise, your chatbots and agent flows can send emails, check inboxes, and manage contacts through visual workflows. MultiMail's gated_send mode is especially important for customer-facing Flowise chatbots, ensuring the bot cannot send emails without human review.

Integrate MultiMail with Flowise using custom tool nodes that call the MultiMail REST API, or through Flowise's built-in HTTP request nodes. Both approaches work with Flowise's chatflow and agentflow patterns.

Built for Flowise developers

Customer-Facing Email Safety

Flowise chatbots often interact directly with customers. MultiMail's gated_send mode ensures that any email sent by the chatbot is reviewed before delivery, critical for maintaining professional communication standards.

Visual Email Flow Builder

Build email-sending chatflows with Flowise's drag-and-drop interface. Non-developers can create workflows that check inboxes, process emails, and send responses — all without writing code.

API-First Integration

Flowise exposes every flow as a REST API. Trigger email processing flows from webhooks, scheduled jobs, or other applications. MultiMail's webhook support can notify your Flowise flow when new emails arrive.

Agentflow Email Tools

Flowise's agentflow pattern supports custom tools. Add MultiMail tools so your agent can decide when to send emails, check inboxes, or manage contacts as part of its reasoning loop.

Low-Code Deployment

Deploy email-capable chatbots quickly with Flowise's hosting options. Combined with MultiMail's managed infrastructure, you get a full email automation stack without managing email servers or complex agent code.


Get started in minutes

Custom MultiMail Tool for Flowise Agentflows
javascript
"cm">// In Flowise, create a Custom Tool with the following configuration:
"cm">// Tool Name: send_email
"cm">// Tool Description: Send an email through MultiMail. In gated_send mode,
"cm">//   the email is queued for human approval before delivery.

"cm">// JavaScript function for the tool:
const fetch = require('node-fetch');

const MULTIMAIL_API = 'https://api.multimail.dev/v1';
const API_KEY = 'mm_live_your_api_key';

async function sendEmail(to, subject, body, mailboxId) {
    const response = await fetch(`${MULTIMAIL_API}/send`, {
        method: 'POST',
        headers: {
            'Authorization': `Bearer ${API_KEY}`,
            'Content-Type': 'application/json'
        },
        body: JSON.stringify({
            mailbox_id: mailboxId,
            to: to,
            subject: subject,
            body: body
        })
    });
    return await response.json();
}

"cm">// Input schema:
"cm">// - to (string): Recipient email address
"cm">// - subject (string): Email subject line
"cm">// - body (string): Email body content
"cm">// - mailboxId (string): Mailbox ID to send from

Create a custom tool node that Flowise agents can use to send emails through MultiMail.

Inbox Checker Tool
javascript
"cm">// Custom Tool: check_inbox
"cm">// Description: Check the MultiMail inbox for recent messages.

const fetch = require('node-fetch');

const MULTIMAIL_API = 'https://api.multimail.dev/v1';
const API_KEY = 'mm_live_your_api_key';

async function checkInbox(mailboxId, limit = 10) {
    const response = await fetch(
        `${MULTIMAIL_API}/mailboxes/${mailboxId}/inbox?limit=${limit}`,
        {
            headers: {
                'Authorization': `Bearer ${API_KEY}`
            }
        }
    );
    return await response.json();
}

"cm">// Input schema:
"cm">// - mailboxId (string): Mailbox ID to check
"cm">// - limit (number, optional): Maximum messages to return (default: 10)

Create a tool that fetches inbox emails for Flowise agents to process.

Trigger Flow from MultiMail Webhook
javascript
"cm">// 1. Deploy your Flowise flow and get the API endpoint:
"cm">//    POST https://your-flowise-instance.com/api/v1/prediction/{flow-id}

"cm">// 2. Create a MultiMail webhook pointing to your Flowise flow:
const fetch = require('node-fetch');

const MULTIMAIL_API = 'https://api.multimail.dev/v1';
const HEADERS = {
    'Authorization': 'Bearer mm_live_your_api_key',
    'Content-Type': 'application/json'
};

"cm">// Register webhook
async function setupWebhook() {
    const resp = await fetch(`${MULTIMAIL_API}/webhooks`, {
        method: 'POST',
        headers: HEADERS,
        body: JSON.stringify({
            url: 'https://your-flowise-instance.com/api/v1/prediction/your-flow-id',
            events: ['email.received'],
            mailbox_id: 'your_mailbox_id'
        })
    });
    console.log(await resp.json());
}

"cm">// 3. When a new email arrives, MultiMail sends a POST to your
"cm">//    Flowise flow with the email data. The flow processes it
"cm">//    through your chatflow/agentflow and sends a reply via
"cm">//    MultiMail's send/reply endpoint.

// 4. Call the Flowise API programmatically:
async function triggerFlow(question) {
    const resp = await fetch(
        'https:"cm">//your-flowise-instance.com/api/v1/prediction/your-flow-id',
        {
            method: 'POST',
            headers: { 'Content-Type': 'application/json' },
            body: JSON.stringify({ question })
        }
    );
    return await resp.json();
}

Use MultiMail's webhook to trigger a Flowise flow when new emails arrive.


Step by step

1

Create a MultiMail Account and API Key

Sign up at multimail.dev, create a mailbox, and generate an API key. Your key will start with mm_live_.

2

Install and Launch Flowise

Install Flowise and open the visual builder in your browser.

bash
npm install -g flowise && npx flowise start
3

Add MultiMail Tools

Create custom tool nodes for send_email and check_inbox that call MultiMail's REST API, or use Flowise's HTTP Request node with MultiMail endpoint URLs.

4

Build Your Flow

Create a chatflow or agentflow that uses your MultiMail tools. Connect an LLM, add your email tools, and configure the system prompt to explain the email context and oversight mode.

5

Test and Deploy

Test your flow in Flowise's built-in chat interface. Review pending emails in the MultiMail dashboard. Deploy the flow as an API endpoint when ready.


Common questions

Can I build email automation in Flowise without writing code?
Mostly yes. Use Flowise's built-in HTTP Request node to call MultiMail API endpoints directly with URL, headers, and body configured visually. For more advanced integration, custom tool nodes require minimal JavaScript. Either way, the workflow logic is built entirely in the drag-and-drop interface.
How do I trigger a Flowise flow when a new email arrives?
Set up a MultiMail webhook that points to your Flowise flow's API endpoint. When a new email arrives in MultiMail, it sends a POST request to your Flowise flow with the email data. The flow processes the email through your chatflow or agentflow and can send a reply via MultiMail.
Should I use chatflows or agentflows for email processing?
Use agentflows when the AI needs to decide which email operations to perform (check inbox, classify, reply). Use chatflows for fixed email processing pipelines where the steps are always the same. Agentflows are more flexible but less predictable; chatflows are deterministic.
Why is gated_send mode important for Flowise chatbots?
Flowise chatbots often face end users who may request inappropriate emails. gated_send mode ensures that every email the chatbot drafts goes through human review before delivery. This is critical for customer-facing deployments where email quality directly impacts brand reputation.
Can I embed a Flowise chatbot with email capabilities in my website?
Yes. Flowise provides embed widgets for websites. Your embedded chatbot can use MultiMail tools to handle email tasks. Users interact via chat, the bot processes email operations via MultiMail, and gated_send ensures appropriate oversight. The MultiMail dashboard serves as the human review interface.

Explore more

The only agent email with a verifiable sender

Email infrastructure built for AI agents. Verifiable identity, graduated oversight, and a 38-tool MCP server. Formally verified in Lean 4.