TL;DR: Apply vendor-supplied mitigations immediately per CISA KEV requirements; if mitigations are unavailable, discontinue use of the affected product. Organizations using certificate autosigning should audit that configuration as a priority.
What Is the Flaw?
CVE-2026-115731 is a deserialization of untrusted data vulnerability (CWE-502) with a secondary code-injection classification (CWE-94) in the report-processing component of Puppet, as used by Puppet Server in Perforce Puppet Enterprise. It carries a CVSS 3.1 score of 9.9 (Critical) with the vector CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H.
The vulnerability exists because the server deserializes data from submitted agent reports without adequate validation. An attacker who can submit a report containing a crafted rich-data type via a PUT request can cause the Puppet Server process to execute arbitrary code at the privilege level of its service user. The scope is marked Changed in the CVSS vector, meaning successful exploitation can affect resources beyond the vulnerable component itself — consistent with the broad administrative reach Puppet Server typically holds across managed infrastructure.
A particularly serious aspect of this vulnerability is its interaction with certificate autosigning. The Puppet certificate authority (CA) can be configured to automatically sign agent certificate signing requests (CSRs). Where this is done without strong verification, an unauthenticated attacker on the network can obtain a CA-signed certificate and then meet the only authentication requirement needed to submit a malicious report — effectively reducing the attack complexity for an unauthenticated attacker to match the already-low bar described in the CVE.
Who and What Is Affected?
The affected product is Perforce Puppet Enterprise, specifically the Puppet Server component and its report-processing code. Consult the vendor's official advisory for the precise list of affected versions and any build numbers, as version boundaries are not enumerated in the currently available NVD entry.
- Puppet Server operators — any organisation running Puppet Enterprise as an infrastructure automation platform is in scope.
- Environments with CA autosigning enabled — these face an elevated risk because an unauthenticated network attacker can obtain the required certificate without prior access.
- Environments where autosigning is disabled or uses strong verification — exploitation still requires a valid CA-signed agent certificate, limiting the attacker pool to entities that already hold such a credential (e.g., a compromised managed node).
Because Puppet Server typically manages configuration across large numbers of nodes, a compromise of the server itself can have cascading effects on the entire managed estate.
Potential Impact
Successful exploitation allows arbitrary code execution on the Puppet Server as the service user of that process. The practical consequences depend on the privilege level of that service user and the scope of Puppet's reach in the environment, but commonly include:
- Full confidentiality loss — access to Puppet catalogs, Hiera data, secrets, and credentials stored on or accessible from the server.
- Full integrity compromise — ability to modify Puppet code, manifests, and modules that are subsequently deployed to all managed nodes.
- Full availability impact — ability to disrupt the Puppet Server and, by extension, automated configuration management across the managed estate.
- Lateral movement — a compromised Puppet Server is a privileged position from which an attacker can push malicious configuration to every node under management.
CISA added this vulnerability to the Known Exploited Vulnerabilities (KEV) catalog on 2026-11-17, confirming active exploitation in the wild. Federal civilian agencies are subject to BOD 26-04 remediation timelines; all other organisations should treat this with equivalent urgency.
What To Do Now
CISA's required action for KEV entries is unambiguous: apply mitigations in accordance with vendor instructions. If mitigations are unavailable, discontinue use of the product. The following steps reflect that guidance and general defensive practice:
- Check the Perforce / Puppet vendor advisory immediately. Obtain the patched version or specific mitigation instructions directly from Perforce. Do not rely on third-party summaries alone for version-specific guidance.
- Audit certificate autosigning configuration. If your Puppet CA is configured to autosign CSRs without strong identity verification, disable or harden that setting as an emergency measure, independent of patching status. This eliminates the unauthenticated exploitation path.
- Review existing CA-signed certificates. Identify all agents currently holding valid CA-signed certificates. Consider revoking certificates for nodes that are decommissioned, unrecognised, or of uncertain provenance.
- Restrict network access to the Puppet Server. Where possible, limit which hosts can submit PUT requests to the Puppet Server's report endpoint at the network layer while patching is in progress.
- Review Puppet Server logs for anomalous report submissions. Look for unexpected PUT requests to report-related endpoints, particularly from unfamiliar agent identities or IP addresses.
- Apply the vendor patch and verify. Once a patch is available and tested, deploy it across all Puppet Server instances. Confirm the installed version matches the vendor's remediated release.
- Refer to CISA BOD 26-04. Federal agencies must follow the patching timeline and forensic triage requirements referenced in the KEV entry. Non-federal organisations should use BOD 26-04 as a remediation benchmark.
How External Testing Surfaces Exposure
Understanding where and how your Puppet infrastructure is externally visible is an important part of assessing your exposure to this vulnerability.
Point-in-time external security assessments can help identify:
- Internet-exposed Puppet Server endpoints — Puppet Server listens on specific TCP ports (commonly 8140). An external scan will reveal whether these ports are reachable from the internet, which dramatically increases risk given the low attack complexity and the unauthenticated path available when autosigning is enabled.
- Service banners and version disclosure — externally visible service responses may indicate the Puppet Server version, allowing an assessor (or an attacker) to determine whether a patched or unpatched version is running.
- Certificate endpoint accessibility — checking whether certificate-signing endpoints are reachable from outside trusted network segments helps gauge the feasibility of the unauthenticated certificate-acquisition path.
It is important to note that an external scan alone cannot confirm that the deserialization vulnerability itself is exploitable without active testing against the target — which would be intrusive. However, confirming that Puppet Server ports are not reachable from untrusted networks is a meaningful, non-intrusive exposure check that can be performed safely.
Internal vulnerability scanning against Puppet infrastructure should also be part of your assessment to verify patch status across all Puppet Server instances.
Sources & further reading: NVD CVE-2026-115731 and the CISA KEV catalog are the authority on affected and fixed versions. Vendor/independent analysis: portal.perforce.com, cisa.gov. This advisory contains no exploit code.
Frequently asked questions
Does an attacker need an existing account or certificate to exploit this vulnerability?
In the general case, an attacker needs a CA-signed agent certificate. However, where the Puppet CA is configured to autosign certificate signing requests without strong verification, an unauthenticated network attacker can obtain such a certificate and then exploit the vulnerability — effectively removing the authentication barrier.
Is this vulnerability being actively exploited?
Yes. CISA added CVE-2026-115731 to the Known Exploited Vulnerabilities (KEV) catalog on 2026-11-17, which confirms active exploitation in the wild. This is the primary authoritative source for that determination.
If I cannot patch immediately, what is the most impactful interim mitigation?
Two interim measures have the highest impact: (1) disable or harden certificate autosigning on your Puppet CA to eliminate the unauthenticated exploitation path, and (2) restrict network access to the Puppet Server report endpoint so that only authorised, known agents can reach it. Neither measure is a substitute for patching, but both significantly raise the bar for exploitation while a patch is being tested and deployed.
Where can I find official remediation guidance?
Consult the official Perforce / Puppet vendor advisory for version-specific patch information. The CISA KEV catalog entry for CVE-2026-115731 (available at cisa.gov/known-exploited-vulnerabilities-catalog) references BOD 26-04 and associated forensic triage requirements. NVD also tracks this CVE at nvd.nist.gov.