# CVE-2026-114728: Roundcube Webmail remote content blocking bypass in CSS FuncIRI handling

Source: https://junglewise.ai/threats/cve-2026-114728-roundcube-webmail-before-1-6-20-and-1-7-x-before-1-7-5-is  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-10-22. Updated: 2026-10-22  
License: CC BY 4.0 with a link back

CVE-2026-114728 · Severity: medium · CVSS 5.3 · Published 2026-10-22

Roundcube Webmail is an email client that runs in a web browser. The vulnerability allows remote content to be loaded from HTML emails without the user's knowledge by bypassing security filters through alternative CSS escape sequences. An attacker could use this to track email opens, steal cookies, or perform other attacks by embedding external content in a specially crafted email.

Technologies: [Roundcube Webmail](https://junglewise.ai/threats/technologies/webmail). Vendors: [Roundcube](https://junglewise.ai/threats/vendors/roundcube).

## Quick answers

**Is CVE-2026-114728 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-114728 being exploited in the wild.

**How severe is CVE-2026-114728?** Medium severity, with a CVSS score of 5.3, published 22 October 2026.

**Which products and versions are affected?** Roundcube Webmail before 1.6.20, 1.7.x before 1.7.5.

**Is there a fix for CVE-2026-114728?** Yes: fixed in Roundcube Webmail 1.6.20, 1.7.5. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

**How is CVE-2026-114728 attacked?** It is attacked over the network, in a web application.

**Am I affected by CVE-2026-114728?** Check whether you run Roundcube Webmail at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## How to fix

- Fixed in Roundcube Webmail 1.6.20, 1.7.5

## Technical details

The rcube_washtml HTML sanitizer fails to detect CSS url() tokens when using alternative escape sequences like \000075rl( or u\72l(. These escaped spellings bypass the content blocking filter that normally prevents remote content in email messages. The vulnerability exists in the handling of FuncIRI attributes and is resolved by normalizing CSS escapes before pattern matching.

## Affected products

- Roundcube Webmail before 1.6.20, 1.7.x before 1.7.5

## Timeline

- 2026-10-22: disclosed
- 2026-10-19: patched

## References

- [https://bugs.debian.org/1151279](https://bugs.debian.org/1151279)
- [https://github.com/roundcube/roundcubemail/commit/2f74267f0e13bed7182b1d2294cc4f3c5c46dcee](https://github.com/roundcube/roundcubemail/commit/2f74267f0e13bed7182b1d2294cc4f3c5c46dcee)

## Related threats

- [Roundcube Webmail improper authentication in multi-host installations](https://junglewise.ai/threats/cve-2026-114732-roundcube-webmail-before-1-6-20-and-1-7-x-before-1-7-5-is)
- [Roundcube Webmail uncontrolled resource consumption in CSS proxy](https://junglewise.ai/threats/cve-2026-114731-roundcube-webmail-before-1-6-20-and-1-7-x-before-1-7-5-is)
- [Roundcube Webmail server-side request forgery filter bypass](https://junglewise.ai/threats/cve-2026-114730-roundcube-webmail-1-6-14-before-1-6-20-and-1-7-0-before-1-7-5-is)
- [Roundcube Webmail content blocking bypass via SVG SMIL](https://junglewise.ai/threats/cve-2026-114729-roundcube-webmail-before-1-6-20-and-1-7-x-before-1-7-5-is)
- [Roundcube Webmail unsafe content passed to ImageMagick](https://junglewise.ai/threats/cve-2026-114727-roundcube-webmail-before-1-6-20-and-1-7-x-before-1-7-5-passes)