Friday, June 12, 2026
Daily digest
The actively exploited Ivanti Sentry max-severity flaw is the story of the day — CISA issued an emergency BOD with a 3-day patch deadline, and exploitation began within 24 hours of disclosure. If Ivanti is anywhere in your environment or supply chain, this requires immediate action.
CISA Issues BOD 26-04: Critical Exploited Flaws Must Be Patched Within 3 Days
CISA published Binding Operational Directive 26-04, requiring Federal Civilian Executive Branch agencies to remediate actively exploited critical vulnerabilities within 72 hours of KEV catalog addition, down from the previous 14-day window for critical flaws. The first vulnerability tagged under the new directive is CVE-2026-10520, an OS command injection flaw in Ivanti Sentry. The directive signals a structural tightening of federal patch cadence expectations.
Why it matters: BOD 26-04 sets a new 72-hour remediation clock for KEV-listed critical CVEs affecting FCEB agencies — a standard that CMMC assessors and DoD primes will likely reference when evaluating contractor patch management practices under NIST 800-171 SI-2. Revisit your patch SLAs now to ensure they can meet this pace.
Max-Severity Ivanti Sentry Flaw (CVE-2026-10520) Exploited Within 24 Hours of Disclosure
A critical OS command injection vulnerability in Ivanti Sentry (CVE-2026-10520) was actively exploited in the wild within 24 hours of public disclosure, prompting CISA to add it to the KEV catalog and order federal agencies to patch by Sunday under BOD 26-04. Dark Reading reports that attackers appear to have pre-mapped Ivanti deployments in anticipation of the release. Ivanti has issued a patch.
Why it matters: If Ivanti Sentry is in your environment or used by any managed service provider touching your network, patch immediately — the 24-hour exploit timeline means vulnerability windows are now measured in hours, not days.
ShinyHunters Exploits Oracle PeopleSoft Zero-Day (CVE-2026-35273) to Breach 100+ Organizations
The ShinyHunters threat group exploited CVE-2026-35273, an unauthenticated remote code execution flaw in Oracle PeopleSoft, to breach over 100 organizations between May 27 and June 9, 2026. Oracle did not publish its advisory until June 10, meaning the bug was a true zero-day during the attack window. Mandiant attributes the campaign to UNC6240 and notes universities were the hardest-hit sector, with at least 40 GB of data stolen from the University of Nottingham alone.
Why it matters: If Oracle PeopleSoft is used anywhere in your org or by contractors sharing your CUI boundary, apply Oracle's mitigation immediately. The unauthenticated RCE vector and the scale of confirmed victims (100+ orgs in under two weeks) make this extremely high priority.
New GreatXML Exploit Bypasses Windows BitLocker via Recovery Partition XML Files
Security researcher Nightmare-Eclipse published a new Windows BitLocker bypass exploit called GreatXML, which leverages maliciously crafted XML files in the Windows recovery partition to circumvent full-disk encryption protections. The researcher describes it as an accidental discovery that took approximately four hours to find, and it follows a separate Microsoft Defender exploit the same researcher dropped the previous day. No CVE or patch has been announced.
Why it matters: BitLocker is a core data-at-rest control for CMMC L2 compliance (NIST 800-171 MP-5 equivalent via SC-28). An unpatched bypass on your Intune-managed Windows 11 fleet could directly undermine a scored practice — monitor for a Microsoft advisory and consider compensating controls around physical access to recovery partitions.
Microsoft Fixes BitLocker Recovery Bug on Windows Server 2025
Microsoft resolved a known issue where Windows Server 2025 devices booted into BitLocker recovery mode after installing the April 2026 security update. The fix is now available and addresses devices that were inadvertently locked out following routine patching.
Why it matters: If you deferred the April 2026 Windows Server 2025 update due to the BitLocker recovery loop issue, the fix is now available — you can resume patching that update ring without risk of lockouts.
LangGraph Flaw Chain Exposes Self-Hosted AI Agents to Remote Code Execution
Researchers disclosed a chain of three now-patched vulnerabilities in LangGraph, an open-source multi-agent AI framework by LangChain, including a critical SQL injection that could lead to remote code execution on self-hosted deployments. All three flaws have been patched in updated LangGraph releases. LangGraph is widely used in enterprise and self-hosted AI agent pipelines.
Why it matters: If your self-hosted AI stack includes LangGraph-based agents, update immediately — an RCE vulnerability in a self-hosted component processing CUI-adjacent data would be a serious CMMC boundary risk.
Agentjacking Attack Tricks AI Coding Agents Into Executing Malicious Code via Sentry
Tenet Security disclosed a new attack class called "Agentjacking" in which adversaries craft fake error reports through Sentry, the open-source error-tracking platform, to manipulate AI coding agents into executing arbitrary code on developer machines. The attack requires no direct access to the developer's environment and exploits the trust AI agents place in tool outputs. No patch exists; the attack is a design-level concern with AI agent architectures.
Why it matters: If developers in your environment use AI coding agents (e.g., GitHub Copilot, Cursor, or custom agents) alongside Sentry for error tracking, this attack vector could result in arbitrary code execution on a dev workstation inside your CUI boundary — review agent permissions and sandbox dev tooling accordingly.
INTERPOL Operation Dismantles Sniper Dz Phishing-as-a-Service Platform, 201 Arrested
INTERPOL's Operation Ramz, conducted between October 2025 and February 2026 across 13 MENA-region countries, resulted in the takedown of Sniper Dz, a decade-long phishing-as-a-service platform, and the arrest of 201 individuals including the platform's primary administrator. Sniper Dz had provided ready-made phishing kits to thousands of criminals. Group-IB supported the operation with threat intelligence.
CISA Weighs Giving Mythos AI Model Access to Scan Federal Agency Networks
White House discussions are reportedly underway to grant CISA access to Mythos, an advanced AI model designed to autonomously detect previously undiscovered cyber vulnerabilities, for the purpose of scanning federal agency networks. No final decision has been announced. The move would represent a significant expansion of AI-assisted offensive-style analysis within the federal cybersecurity posture.
Ansible Automation Platform 2.7 Introduces Visual Execution Environment Builder
Red Hat released Ansible Automation Platform 2.7, featuring a new visual execution environment builder and unified content discovery experience within the automation portal. The update is intended to reduce time spent manually managing dependencies, tracking content collections across repos, and configuring execution environments. The release is generally available now.
Why it matters: If you run Ansible Automation Platform for configuration management across your Nutanix AHV or AWS GovCloud workloads, the new visual EE builder can reduce the overhead of maintaining compliant, pinned execution environments — relevant for keeping your automation stack auditable under CMMC CM controls.
Terraform MCP Server Reaches General Availability
HashiCorp announced that the Terraform MCP (Model Context Protocol) server has reached general availability at version 1.0, enabling AI assistants to interact with Terraform workflows for infrastructure management. The server supports flexible deployment options and is designed to enforce consistent infrastructure configurations across organizations. It integrates with MCP-compatible AI tools.
NITAAC Announces Sunset of All Government-Wide IT Contracting Vehicles by End of 2028
The NIH Information Technology Acquisition and Assessment Center (NITAAC) announced it will cease all program functions, including its government-wide acquisition vehicles, after December 31, 2028. NITAAC previously managed the CIO-SP3 vehicle, a $50 billion IT contracting program heavily used by federal agencies and DoD contractors. Agencies and contractors using NITAAC vehicles will need to transition to alternative procurement mechanisms.
Why it matters: If your organization holds or plans to compete for contracts under CIO-SP3 or other NITAAC vehicles, begin transition planning now — the 2028 sunset gives a defined window but procurement lead times for DoD IT work mean action is needed well before the deadline.