News

Cisco Email Gateway Writes Root Files From Email

Louis Stowasser
Louis Stowasser
Thursday 15 October 2026
Cisco Email Gateway Writes Root Files From Email
Cisco Email Gateway Writes Root Files From Email

Cisco Secure Email Gateway sits in front of an organisation’s mail system, inspecting inbound and outbound messages for phishing, spam, malicious links and attachments, data loss and other policy violations. It is deployed as physical or virtual appliances and in hosted or hybrid arrangements, by organisations using Microsoft 365, Google Workspace or Exchange—especially larger and distributed enterprises that centralise mail filtering.

CVE-2026-76519 is an arbitrary-file-write flaw in the product’s Advanced Malware Protection (AMP) File Reputation processing. That feature evaluates attachment reputation through Cisco’s cloud service. When it is enabled on an incoming or outgoing mail policy, a specially constructed attachment can abuse relative path traversal: input intended to identify or store an attachment is handled in a way that lets it escape its expected directory and write somewhere else on the underlying appliance.

The attacker needs no account, prior access or user interaction. They need only be able to send the crafted email through a vulnerable gateway with the feature enabled. Cisco says successful exploitation can write arbitrary files and execute commands with root privileges, turning routine handling of untrusted email into a route to control of the appliance.

Exploitation makes an edge-mail flaw urgent

Cisco confirmed active exploitation during October 2026. CVE-2026-76519 entered CISA’s Known Exploited Vulnerabilities catalogue on October 14, with an October 17 federal remediation deadline; the catalogue records ransomware use as unknown. Cisco has not publicly identified an actor, campaign, victim, payload or the scale of compromise.

That limited detail should not be mistaken for a limited consequence. A mail gateway is deliberately exposed to messages from outside the organisation and commonly has privileged access to the systems and services needed to process mail. Cisco also warns that an intruder with root access could steal cluster SSH private keys, potentially creating a path toward other members of a gateway cluster.

A GitHub repository claims to be a proof of concept, but its publicly available contents were only a README when reviewed on October 15. No exploit script, requirements or executable source was available there. In short, a claimed PoC repository exists, but no working public exploit code, Metasploit module, Exploit-DB entry or independently validated PoC has surfaced.

Fix the scanner component, then look for evidence

A device is exposed when it runs vulnerable AsyncOS, has AMP File Reputation assigned to a mail policy, and uses Content Scanner Tools earlier than 25.3.0.5525. Cisco distributes the correction in Content Scanner Tools 25.3.0.5525 and later; this update does not require an AsyncOS upgrade or appliance restart. First fixed platform releases are AsyncOS 15.5.5-014 and 16.5.0-780; Cisco lists no fixed 16.0 release and says administrators should migrate. There is no workaround.

Teams should verify that the scanner-tools update actually landed, rather than assuming automatic updates are enabled. Cisco Secure Email Cloud Gateway has already received the managed fix. Secure Email and Web Manager and Secure Web Appliance are not vulnerable to CVE-2026-76519.

Treat patching as only the first task where exposure may have existed. Review the gateway’s system_logs for Cisco’s compromise indicators, but preserve and examine external firewall and network logs too: root-level attackers may be able to remove local evidence, and unexpected uploads or downloads involving the gateway deserve attention. For clusters, assess credentials and peer appliances as part of incident response.

For mail-security operators, this is a case where an unauthenticated sender can reach a root-level outcome through a normal business workflow. Use SecAlerts to monitor the actual software stack and alert on new vulnerabilities affecting products the organisation runs, then pair that visibility with prompt scanner updates and evidence-led compromise checks.

Contact

SecAlerts Pty Ltd.
132 Wickham Terrace
Fortitude Valley,
QLD 4006, Australia
info@secalerts.co
By using SecAlerts services, you agree to our services end-user license agreement. This website is safeguarded by reCAPTCHA and governed by the Google Privacy Policy and Terms of Service. All names, logos, and brands of products are owned by their respective owners, and any usage of these names, logos, and brands for identification purposes only does not imply endorsement. If you possess any content that requires removal, please get in touch with us.
© 2026 SecAlerts Pty Ltd.
ABN: 70 645 966 203, ACN: 645 966 203