# CVE-2026-46449: Akamai Guardicore Platform Agent command injection in wapptunneld

Source: https://junglewise.ai/threats/cve-2026-46449-the-akamai-guardicore-platform-agent-gpa-before-7-4-for-macos-is  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-10-14. Updated: 2026-10-16  
License: CC BY 4.0 with a link back

CVE-2026-46449 · Severity: medium · CVSS 6.7 · Published 2026-10-14

Akamai's Guardicore Platform Agent (GPA) is endpoint security software that manages network access on macOS devices. A local attacker can rename network interfaces to inject malicious commands that execute with root privileges when the user logs out from their identity provider, potentially taking full control of the system.

Technologies: [Akamai Guardicore Platform Agent](https://junglewise.ai/threats/technologies/guardicore-platform-agent). Vendors: [Akamai](https://junglewise.ai/threats/vendors/akamai).

## Quick answers

**Is CVE-2026-46449 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-46449 being exploited in the wild.

**How severe is CVE-2026-46449?** Medium severity, with a CVSS score of 6.7, published 14 October 2026.

**Which products and versions are affected?** Akamai Guardicore Platform Agent before 7.4.

**Is there a fix for CVE-2026-46449?** Yes: fixed in 7.4. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

**How is CVE-2026-46449 attacked?** It is attacked with local access, in a network appliance.

**Am I affected by CVE-2026-46449?** Check whether you run Akamai Guardicore Platform Agent 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 7.4

## Technical details

The wapptunneld service in GPA before 7.4 fails to validate network interface names before passing them to system utilities, allowing command injection. An unprivileged local user can exploit this by modifying network interface names via System Settings; when the "Log out from your IdP" action is triggered, the unsanitized interface name executes arbitrary commands as root.

## Affected products

- Akamai Guardicore Platform Agent before 7.4

## Timeline

- 2026-10-14: disclosed

## References

- [https://www.akamai.com/newsroom/security-advisories/guardicore-platform-agent-7-4](https://www.akamai.com/newsroom/security-advisories/guardicore-platform-agent-7-4)

## Related threats

- [Akamai Guardicore Platform Agent improper JWT validation](https://junglewise.ai/threats/cve-2026-50218-akamai-gpa-guardicore-platform-agent-client-on-all-platforms)
- [Akamai Guardicore Platform Agent use-after-free in WFP callout driver](https://junglewise.ai/threats/cve-2026-50217-a-use-after-free-vulnerability-in-the-windows-wfp-callout-driver)
- [Akamai Guardicore Platform Agent integer underflow in DNS proxy](https://junglewise.ai/threats/cve-2026-50216-an-integer-underflow-exists-in-the-dns-proxy-component-of-the)
- [Akamai Guardicore Platform Agent use-after-free in WFP driver](https://junglewise.ai/threats/cve-2026-50215-a-use-after-free-exists-in-the-windows-wfp-packet-capture-driver)
- [Akamai Guardicore Platform Agent local privilege escalation](https://junglewise.ai/threats/cve-2026-34354-akamai-guardicore-platform-agent-local-privilege-escalation)