CYBER_THREATCAST
$ briefing --date=

CYBER THREATCAST

CYBER THREAT INTELLIGENCE BRIEFING

Wednesday, November 18, 2026|AFTERNOON EDITION|15:39 TR (12:39 UTC)|192 Signals|15 Sectors
CISA confirmed attackers are exploiting CVE-2026-115731 to execute code on vulnerable Perforce Puppet servers. The agency added the flaw to its Known Exploited Vulnerabilities catalog, a public proof of concept is available, and patched versions have been released.
CVE-2026-115731 allows a client holding a Puppet CA-signed agent certificate to submit a malicious report, trigger unsafe deserialization and run arbitrary code with the Puppet service user’s privileges. The certificate requirement limits the path, but successful exploitation gives attackers a powerful foothold in management infrastructure.
A public proof of concept targeting OpenVox adds urgency. Organizations running Puppet Core, Puppet Enterprise or OpenVox should prioritize patched releases and investigate whether CA-signed client certificates could have been abused.

Editorial: Recommended Actions

01
PRIORITY
Patch vulnerable Perforce Puppet Enterprise, Puppet Core, Puppet Server, and OpenVox deployments immediately using the available fixed versions. CISA has confirmed active exploitation of CVE-2026-115731, and a public proof of concept exists; any holder of a Puppet CA-signed client certificate can submit a malicious report and execute arbitrary code with service-user privileges.
02
PRIORITY
Upgrade Apache HTTP Server 2.4.55 through 2.4.70 to version 2.4.71 immediately wherever mod_http2 and HTTP/2 are enabled. If rapid patching is impossible, temporarily disable HTTP/2; CVE-2026-119618 is reportedly under active exploitation and can corrupt heap memory, potentially enabling unauthenticated remote code execution. HTTP/1.1-only servers are not exposed to this HTTP/2-specific flaw.
03
PRIORITY
Update ConnectWise ScreenConnect host clients to version 26.6.5 or later and inspect affected MSP environments for malicious guest file-execution requests, rogue ScreenConnect clients, batch files contacting a Telegram bot, and startup-folder persistence scripts. Attackers exploited CVE-2026-84869 against 42 MSP servers and generated at least 688,000 malicious requests while attempting to compromise technician systems.
04
PRIORITY
Update on-premises N-able N-central servers, Agents, and Probes immediately, and verify that legacy APIs are disabled. Attackers used public proof-of-concept code within hours of disclosure; the 4EverAble2Shell chain combines Apache Axis remote code execution with an N-central weakness to run shell scripts as root and deploy cryptocurrency miners.
05
PRIORITY
Treat every system that installed Microsoft agent-framework versions 1.26.1 or 1.26.2 as compromised: remove the packages, revoke tokens, rotate credentials, and review audit logs. The malicious PyPI releases executed through Python .pth startup hooks and harvested stored secrets plus fresh Azure managed-identity, federation, and GitHub Actions OIDC tokens, exposing developers, CI runners, and Azure workloads.

Field Signals

Real-time intelligence from X/Twitter
$ scanning feeds_

Sector Intelligence

Cyber Threatcast is generated by an autonomous AI intelligence pipeline. All assessments are algorithmically derived.

Published by halilozturkci.com