CVE-2026-34724: 
NixOS vulnerability analysis and mitigation

CVE-2026-34724 is a server-side template injection (SSTI) vulnerability in Zammad, a web-based open source helpdesk/customer support system, that leads to remote code execution (RCE) via the AI Agent feature. It affects Zammad version 7.0.0 and was disclosed on April 8, 2026, with a patch released in version 7.0.1. The vulnerability carries a CVSS v3.1 base score of 7.2 (High) and a CVSS v4.0 base score of 8.7 (High) (GitHub Advisory, Red Hat).

The vulnerability is rooted in improper neutralization of special elements used in a template engine (CWE-1336) and improper control of code generation (CWE-94). Specifically, the type_enrichment_data configuration parameter of the AI Agent feature is processed through a template engine without adequate sanitization, allowing injected template expressions (e.g., ERB-like sequences) to be evaluated server-side. Exploitation requires an attacker to control or influence the type_enrichment_data field, which is typically accessible only through high-privilege administrative configuration. A public proof-of-concept exploit has been published at https://github.com/0xBlackash/CVE-2026-34724 (GitHub Advisory).

Successful exploitation results in remote code execution on the Zammad application server in the context of the application process, yielding full confidentiality, integrity, and availability impact. An attacker could exfiltrate sensitive customer support data and tickets, modify or delete support records, and potentially disrupt or shut down the service. While the requirement for high-privilege administrative access limits the attack surface, a compromised admin account or insider threat could leverage this vulnerability for complete server takeover (GitHub Advisory).

A public proof-of-concept exploit has been published on GitHub at https://github.com/0xBlackash/CVE-2026-34724, attributed to researcher 0xBlackash. As of the available intelligence, there is no confirmed evidence of active in-the-wild exploitation. The EPSS score is approximately 0.043%, reflecting a currently low probability of exploitation in the near term. The vulnerability is not listed in the CISA Known Exploited Vulnerabilities (KEV) catalog (GitHub Advisory).

1. Reconnaissance: Identify Zammad instances running version 7.0.0 using internet scanning tools (e.g., Shodan, Censys) or by checking the Zammad version endpoint.
2. Obtain administrative credentials: Acquire high-privilege (admin) credentials through phishing, credential stuffing, or insider access, as exploitation requires administrative access to the AI Agent configuration.
3. Access AI Agent configuration: Log in to the Zammad admin panel and navigate to the AI Agent settings where type_enrichment_data can be configured.
4. Inject malicious template payload: Insert a server-side template injection payload (e.g., an ERB expression such as <%= system('id') %> or a reverse shell command) into the type_enrichment_data field.
5. Trigger template evaluation: Save the configuration and trigger the AI Agent functionality to process the type_enrichment_data field, causing the template engine to evaluate the injected payload.
6. Achieve RCE: The injected code executes in the context of the Zammad application process, enabling arbitrary command execution, data exfiltration, or further lateral movement (GitHub Advisory).

• Logs: Zammad application logs showing unexpected errors or output from template engine evaluation; admin audit logs recording modifications to AI Agent or type_enrichment_data configuration by unusual accounts or at unusual times.
• File System: Unexpected files written to the Zammad application directory or temporary directories by the application process; new cron jobs or scripts created under the Zammad service account.
• Network: Outbound connections from the Zammad server to unknown external IPs or domains, particularly following AI Agent configuration changes; reverse shell traffic originating from the application process.
• Process: Unusual child processes spawned by the Zammad Ruby/Rails application process (e.g., bash, sh, curl, wget, python) (GitHub Advisory).

Upgrade Zammad to version 7.0.1 or later, which contains the fix for this vulnerability. As temporary risk reduction measures prior to patching: restrict and audit privileged (admin) accounts to ensure only trusted users can configure AI Agents; audit AI::Agent records and type_enrichment_data fields for suspicious content such as ERB-like sequences; and consider disabling or limiting the AI Agent feature entirely until the patch is applied (GitHub Advisory).

The vulnerability was reported by security researcher k0x1c and disclosed via GitHub Security Advisories by Zammad maintainer fliebe92. Social media activity was observed on Bluesky and Twitter/X (via FOFA bot), indicating moderate community awareness. No major vendor statements beyond the official advisory or significant media coverage have been identified at this time (GitHub Advisory).