🔍 Read the full analysis: What We Know About The PHP2Shell Chain And Its AI Assistance on ThorstenMeyerAI.com
Get business pricing on monitors, keyboards and dev gear
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
Dream says an attacker chained two PHP2Shell flaws with the previously disclosed Fragnesia Linux kernel bug, gaining root access on 24 internet-exposed PAN-OS firewalls and reaching a U.S. government-affiliated network. Dream found AI coding and security tools on the attacker’s servers but said it has no direct evidence they created the exploit or scripts. Palo Alto Networks says Unit 42 independently verified the attacker’s tools and results.
An attacker gained root access to 24 PAN-OS firewalls and used one organization’s compromised firewalls to reach an internal server on a U.S. government-affiliated network, according to a report by Israeli security company Dream. Dream says the attacker chained the newly patched PHP2Shell flaws with a Linux kernel vulnerability disclosed months earlier; the incident has also drawn attention to AI tools found on the attacker’s servers, though their role in building the attack is unproven.
Dream’s account, published October 26 and updated October 29, says the attacker ran an automated batch against 240 firewalls on October 23, roughly two days after PHP and Palo Alto Networks published security information about PHP2Shell. Dream reports that 24 firewalls were rooted. The targets were in the United States, Japan, Taiwan and South Korea. Palo Alto Networks says its Unit 42 researchers independently verified the attacker’s tools and results.
The attack combined CVE-2026-109618 and CVE-2026-109619, known as PHP2Shell, with Fragnesia (CVE-2026-46300), a Linux kernel privilege-escalation flaw made public in May. Dream says PHP2Shell provided command execution on the firewall as a low-privilege account; Fragnesia then enabled the attacker to gain root access. Dream’s report says 22 of the 24 compromised firewalls were left with a hidden web shell.
Dream says two compromised firewalls belonged to a U.S. government-affiliated organization. The attacker used one as a route to the organization’s internal firewall-management server, taking over a 25th server, then targeted Active Directory, which manages user identities and access. Dream’s account describes persistent access, but the supplied material does not establish the extent of any resulting data theft or operational disruption.
Infographic 01 · PHP2Shell on Palo Alto firewalls
Two days from patch to root.
PHP fixed the bug on 21 October. Fixed firewall software for the attacked branches arrived almost a week later. The attack ran in between.
- 13–23 MayFragnesia goes public, kernel fix followsLinux privilege-escalation bug with a public exploit; CVE-2026-46300.
- 8 OctAttacker sets up an AI platformaiscan configured with DeepSeek and Kimi, per Dream’s timeline.
- 21 OctPHP fixes PHP2ShellPHP 8.5.12, 8.4.27, 8.3.36, 8.2.35. Palo Alto advisory the same day; PAN-OS fixes “upcoming”.
- 23 Oct, evening UTCBatch run: 240 targets, 24 rootedFound via device search engine, filtered to US, Japan, Taiwan, South Korea.
- 24 OctInto a government-affiliated networkTunnel, persistence, internal management server rooted (no. 25), Active Directory attacked.
- 26 OctDream publishes · CISA lists PHP2ShellFederal deadline 29 Oct.
- 27–29 OctFixed PAN-OS builds for 12.2 and 12.1CISA lists Fragnesia on 28 Oct; Unit 42 confirms Dream’s results on 29 Oct.
The fix for PHP existed. The fix for the firewall did not, yet. The open management page was the only door that mattered.Thorsten Meyer · my reading
Sources: Dream Research, 26 Oct 2026 (updated 29 Oct); php.net, 21 Oct 2026; Palo Alto Networks advisory (updated 29 Oct 2026); CISA KEV catalog, version 29 Oct 2026; oss-security, 13 May 2026; Unit 42, 29 Oct 2026. Day count: Thorsten Meyer.
Powered by Thorsten Meyer AI
Exposed Management Pages Enabled Access
The reported intrusion shows how an internet-reachable firewall management interface can become a path into the systems it is meant to protect. In this case, the reported chain moved from a perimeter device to an internal server responsible for managing firewalls, followed by activity against Active Directory. That creates potential risks beyond the initially compromised devices, although the available account does not confirm what information, if any, was accessed or taken.
The timing also matters. PHP had released fixes on October 21, but Dream says Palo Alto Networks’ corrected PAN-OS builds for the affected 12.1 and 12.2 branches did not arrive until October 27–29. The vendor’s October 21 advisory identified the issue while fixes were still forthcoming. During that period, keeping management interfaces off the public internet was the stated protective measure. The incident underscores that patch availability and patch deployment are different, and that an older, public kernel flaw can become important when chained with a newer vulnerability.
AI is part of the story because Dream found tools associated with it in the attacker’s environment. But the evidence described does not show that AI generated the exploit or directed the intrusion. The confirmed concern is the reported compromise and the combination of exposed management access, vulnerable software and an additional privilege-escalation flaw—not a proven autonomous AI attack.
Products Worth Considering
LG 34GX90SA-W 34-inch Ultragear WQHD OLED Curved Gaming Monitor
LG UltraGear 27-Inch QHD OLED Gaming Monitor, 240Hz, 0.03ms, G-SYNC Compatible, FreeSync Premium Pro, HDMI 2.1
Amazon Kindle Scribe Colorsoft 64GB (1st Generation)
As an affiliate, we earn on qualifying purchases.
From PHP Fixes to Firewall Patches
PHP2Shell refers to two vulnerabilities in the PHP code included with PAN-OS, Palo Alto Networks’ firewall operating system. Dream says the firewalls used PHP 8.0.30, a version bundled with PAN-OS. The PHP project released updated versions on October 21 and urged users to update. Palo Alto Networks issued its advisory the same day, with fixes for the affected PAN-OS branches to follow.
The second part of the chain, Fragnesia, was publicly disclosed in May 2026 and had a kernel fix and a public exploit, according to the source account. Dream says the attacker first identified roughly 19,000 PAN-OS devices with internet-facing management pages, obtained version information from about 5,000, then narrowed the pool to 240 targets based on country and software version. These are figures attributed to Dream’s analysis, not a count of successful compromises.
Dream’s timeline says the attacker configured an offensive-security platform called aiscan on October 8, with DeepSeek and Kimi models. Researchers found Claude Code installed on the attacker’s servers as well. Dream says it found no direct evidence that those AI tools wrote the exploit or scripts and assessed that the operation did not appear fully autonomous. CISA added both PHP2Shell CVEs to its Known Exploited Vulnerabilities catalog on October 26 and set a three-day deadline for federal agencies, according to the supplied material.
““no direct evidence” that AI tools wrote the exploit or the scripts”
— Dream
AI’s Role and Intrusion Impact
The evidence described by Dream does not establish whether AI generated any exploit code, wrote attack scripts, or materially changed the operation’s speed or success. The presence of Claude Code, DeepSeek and Kimi configurations shows that the tools were available in the attacker’s environment; it does not prove how they were used. Dream explicitly says it has no direct evidence tying them to the exploit or scripts.
The supplied reporting also leaves the full impact unclear. It says the attacker reached the internal firewall-management server and targeted Active Directory, but does not specify whether accounts were altered, data was removed, or the organization suffered lasting disruption. The source says attacks on Active Directory followed, without detailing their outcome. The number of affected devices beyond the reported 24 rooted firewalls and one internal server is also not established here.
Dream reports that the first government-affiliated firewall crashed shortly after the attack began, likely because of about 900 MB of attack traffic. That explanation is presented as likely, not confirmed. The source does not identify the organization, provide a complete victim list, or describe whether every compromised firewall was remediated.
Patch, Restrict, and Investigate
Organizations running affected PAN-OS versions should consult Palo Alto Networks’ current advisories and apply the fixed releases for their branch. The source account says fixes for the affected 12.1 and 12.2 branches shipped between October 27 and 29. Where management access is not needed from the public internet, administrators should restrict it to trusted networks, consistent with the vendor’s interim guidance while fixes were pending.
Operators should also investigate exposed management interfaces and affected devices for signs of compromise, including the persistence described by Dream, and review access to internal management systems and Active Directory. CISA’s addition of both PHP2Shell CVEs to its exploited-vulnerability catalog puts the issue on federal agencies’ remediation schedule; the supplied source says agencies were given three days.
Further clarity depends on incident investigations and any additional updates from Dream, Palo Alto Networks, the affected organization or government agencies. Until then, the intrusion’s full impact and the specific contribution of AI tools remain unconfirmed.
Key Questions
What is PHP2Shell?
PHP2Shell is the name used for two vulnerabilities, CVE-2026-109618 and CVE-2026-109619, affecting PHP bundled with PAN-OS. Dream says the flaws enabled command execution on targeted firewalls.
How did the attacker gain root access?
According to Dream, the attacker chained PHP2Shell with Fragnesia (CVE-2026-46300), a Linux kernel privilege-escalation flaw disclosed in May. Dream says the combination moved access from a low-privilege account to root.
Did AI build or run the attack?
That is not established. Dream found Claude Code and configurations for DeepSeek and Kimi on the attacker’s servers, but said it had no direct evidence that AI wrote the exploit or scripts and did not consider the operation fully autonomous.
How many systems were reported compromised?
Dream reports 24 rooted PAN-OS firewalls and one internal firewall-management server, for 25 servers in total. The report does not establish the complete impact on data or accounts.
What should PAN-OS administrators do?
Check Palo Alto Networks’ advisories and apply the fixed release for the relevant PAN-OS branch. Restrict management interfaces from public access where possible, and investigate affected devices and connected identity systems for signs of compromise.
Source: ThorstenMeyerAI.com
