Withdrawn Advisory: marked cross-site scripting vulnerability

Versions 0.3.2 and earlier of marked are affected by a cross-site scripting vulnerability even when sanitize:true is set.

Inefficient Regular Expression Complexity in marked

Impact

What kind of vulnerability is it?

Denial of service.

The regular expression inline.reflinkSearch may cause catastrophic backtracking against some strings. PoC is the following.

import * as marked from 'marked';

console.log(marked.parse(`[x]: x

\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](\\[\\](`));

Who is impacted?

Anyone who runs untrusted markdown through marked and does not use a worker with a time limit.

Patches

Has the problem been patched?

Yes

What versions should users upgrade to?

4.0.10

Workarounds

Is there a way for users to fix or remediate the vulnerability without upgrading?

Do not run untrusted markdown through marked or run marked on a worker thread and set a reasonable time limit to prevent draining resources.

References

Are there any links users can visit to find out more?

• https://marked.js.org/using_advanced#workers
• https://owasp.org/www-community/attacks/Regular_expression_Denial_of_Service_-_ReDoS

For more information

If you have any questions or comments about this advisory:

• Open an issue in marked

Marked vulnerable to XSS from data URIs

marked version 0.3.6 and earlier is vulnerable to an XSS attack in the data: URI parser.

Cross-Site Scripting in marked

Versions 0.3.7 and earlier of marked unescape only lowercase while owsers support both lowercase and uppercase x in hexadecimal form of HTML character entity

Multiple Content Injection Vulnerabilities in marked

Versions 0.3.0 and earlier of marked are affected by two cross-site scripting vulnerabilities, even when sanitize: true is set.

The attack vectors for this vulnerability are GFM Codeblocks and JavaScript URLs.

Recommendation

Upgrade to version 0.3.1 or later.

VBScript Content Injection in marked

Versions 0.3.2 and earlier of marked are affected by a cross-site scripting vulnerability even when sanitize:true is set.

Proof of Concept ( IE10 Compatibility Mode Only )

[xss link](vbscript:alert(1))

will get a link

<a href="vbscript:alert(1)">xss link</a>

Recommendation

Update to version 0.3.3 or later.

Moderate severity vulnerability that affects marked

Withdrawn

This advisory has been withdrawn, per NVD: This candidate was withdrawn by its CNA. Further investigation showed that it was not a security issue.

Original Description

A Regular expression Denial of Service (ReDoS) vulnerability in the file marked.js of the marked npm package (tested on version 0.3.7) allows a remote attacker to overload and crash a server by passing a maliciously crafted string.

Regular Expression Denial of Service in marked

Versions 0.3.3 and earlier of marked are affected by a regular expression denial of service ( ReDoS ) vulnerability when passed inputs that reach the em inline rule.

Recommendation

Update to version 0.3.4 or later.

Inefficient Regular Expression Complexity in marked

Impact

What kind of vulnerability is it?

Denial of service.

The regular expression block.def may cause catastrophic backtracking against some strings. PoC is the following.

import * as marked from "marked";

marked.parse(`[x]:${' '.repeat(1500)}x ${' '.repeat(1500)} x`);

Who is impacted?

Anyone who runs untrusted markdown through marked and does not use a worker with a time limit.

Patches

Has the problem been patched?

Yes

What versions should users upgrade to?

4.0.10

Workarounds

Is there a way for users to fix or remediate the vulnerability without upgrading?

Do not run untrusted markdown through marked or run marked on a worker thread and set a reasonable time limit to prevent draining resources.

References

Are there any links users can visit to find out more?

• https://marked.js.org/using_advanced#workers
• https://owasp.org/www-community/attacks/Regular_expression_Denial_of_Service_-_ReDoS

For more information

If you have any questions or comments about this advisory:

• Open an issue in marked

Sanitization bypass using HTML Entities in marked

Affected versions of marked are susceptible to a cross-site scripting vulnerability in link components when sanitize:true is configured.

Proof of Concept

This flaw exists because link URIs containing HTML entities get processed in an abnormal manner. Any HTML Entities get parsed on a best-effort basis and included in the resulting link, while if that parsing fails that character is omitted.

For example:

A link URI such as javascript&#x58document;alert(1) Renders a valid link that when clicked will execute alert(1).

Recommendation

Update to version 0.3.6 or later.

Content injection in marked

Versions 0.3.7 and earlier of marked When mangling is disabled via option mangle don't escape target href. This allow attacker to inject arbitrary html-event into resulting a tag.

Regular Expression Denial of Service in marked

Affected versions of marked are vulnerable to a regular expression denial of service.

The amplification in this vulnerability is significant, with 1,000 characters resulting in the event loop being blocked for around 6 seconds.

Recommendation

Update to version 0.3.9 or later.

Sanitization bypass using HTML Entities

marked is an application that is meant to parse and compile markdown.

Due to the way that marked parses input, specifically HTML entities, it's possible to bypass marked's content injection protection (sanitize: true) to inject a javascript: URL.

This flaw exists because &#xNNanything; gets parsed to what it could and leaves the rest behind, resulting in just anything; being left.

For example:

If a malicious user could provide this input to the application javascript&#x58document;alert(1) resulting in a valid link, that when a user clicked it would execute alert(1).

Multiple Content Injection Vulnerabilities

Marked comes with an option to sanitize user output to help protect against content injection attacks.

sanitize: true

Even if this option is set, marked is vulnerable to content injection in multiple locations if untrusted user input is allowed to be provided into marked and that output is passed to the browser.

Injection is possible in two locations

• gfm codeblocks (language)
• javascript url's

Regular Expression Denial of Service

Marked 0.3.3 and earlier is vulnerable to regular expression denial of service (ReDoS) when certain types of input are passed in to be parsed.

The Regular expression Denial of Service (ReDoS) is a Denial of Service attack, that exploits the fact that most Regular Expression implementations may reach extreme situations that cause them to work very slowly (exponentially related to input size). An attacker can then cause a program using a Regular Expression to enter these extreme situations and then hang for a very long time. [1]

Marked's catastrophic backtracking issue for the em inline rule has now been patched in 0.3.4.

VBScript Content Injection

Marked 0.3.2 and earlier is vulnerable to content injection even when sanitize: true is enabled.

[xss link](vbscript:alert(1))

will get a link

<a href="vbscript:alert(1)">xss link</a>

this script does not work in IE 11 edge mode, but works in IE 10 compatibility view.