← CYBERDUDEBIVASH SENTINEL APEX SA-2026-0002
CRITICAL TLP:CLEAR HIGH CONFIDENCE CVE-2026-0257

Palo Alto Networks PAN-OS GlobalProtect Authentication Bypass (CVE-2026-0257) — CISA KEV-Listed, Confirmed Ransomware-Campaign Use, Federal Deadline Long Overdue

Published 2026-07-28 · Sentinel APEX Intelligence Division · v1.0

Executive Summary

An authentication bypass vulnerability in the GlobalProtect portal and gateway of Palo Alto Networks PAN-OS (CVE-2026-0257) allows an attacker to bypass authentication and establish an unauthorized VPN connection when Authentication Override cookies are enabled with a specific certificate configuration. NVD rates this CVSS 9.1 (the vendor's own advisory self-rates it 7.8 — see Verified Facts for the discrepancy). EPSS places it in the 99.8th percentile of all scored vulnerabilities — meaning fewer than 2 in 1,000 scored CVEs are more likely to see real-world exploitation. CISA added it to the Known Exploited Vulnerabilities catalog on 2026-05-29 with a federal remediation deadline of 2026-06-01 — now roughly eight weeks overdue as of this report's date. Palo Alto Networks' own advisory confirms limited in-the-wild exploitation against unpatched, unmitigated devices. CISA's KEV catalog separately flags this CVE with knownRansomwareCampaignUse: Known. Four PAN-OS release branches (10.2, 11.1, 11.2, 12.1) and two Prisma Access versions are affected; patches and two documented workarounds exist.

Executive Risk Snapshot

DimensionRatingConfidence
SeverityCRITICAL (CVSS 9.1, NVD) — vendor's own advisory self-rates 7.8/HIGH; see Verified FactsHIGH — both ratings independently confirmed via live queries, discrepancy documented rather than resolved by picking one
Exploitation statusConfirmed — vendor's own words: "limited exploit attempts on unpatched PAN-OS devices without mitigations applied"HIGH — vendor advisory, not third-party claim
ExposureFour PAN-OS branches (10.2/11.1/11.2/12.1) + Prisma Access 10.2/11.2 below specific patched buildsHIGH — vendor-published version ranges
Ransomware associationCISA KEV catalog: knownRansomwareCampaignUse: Known (no specific group named by CISA)MEDIUM — authoritative but non-specific; see Intelligence Gaps for the unverified "Qilin" claim
UrgencyFederal deadline (2026-06-01) passed roughly 8 weeks before this reportHIGH
OverallCRITICALHIGH

Why This Matters

Patch guidance for this CVE, read alone, misses half the exposure. Palo Alto Networks' advisory documents two independent remediation paths: apply the fix, or apply either of two workarounds (use a dedicated certificate for Authentication Override cookies, or disable Authentication Override entirely). [Analyst Assessment] (MEDIUM CONFIDENCE) — neither the vendor advisory nor CISA's KEV entry states whether patching alone retroactively terminates a VPN session or access grant that was already established via the authentication bypass before the patch was applied; architecturally, closing an authentication flaw does not itself revoke a session that flaw already created. An organization that patches today without also reviewing existing GlobalProtect sessions and access-control-list entries for anomalies predating the patch may be closing the door after already-established, unauthorized access remains inside it. This is an inference from how authentication-bypass classes of vulnerability generally behave, not a claim any cited source makes explicitly about this specific CVE — stated at MEDIUM CONFIDENCE for that reason.

Verified Facts — Vulnerability Fact Sheet

FieldValueSource
CVE IDCVE-2026-0257NVD
Vulnerability namePalo Alto Networks PAN-OS Authentication Bypass VulnerabilityCISA KEV
CWECWE-565: Reliance on Cookie without Validation and Integrity CheckingCISA KEV
Affected products / versionsPAN-OS 12.1 < 12.1.4-h6 and < 12.1.7; PAN-OS 11.2 < 11.2.4-h17, < 11.2.7-h14, < 11.2.10-h7, < 11.2.12; PAN-OS 11.1 < 11.1.4-h33, < 11.1.6-h32, < 11.1.7-h6, < 11.1.10-h25, < 11.1.13-h5, < 11.1.15; PAN-OS 10.2 < 10.2.7-h34, < 10.2.10-h36, < 10.2.13-h21, < 10.2.16-h7, < 10.2.18-h6; Prisma Access 11.2.0 < 11.2.7-h13; Prisma Access 10.2.0 < 10.2.10-h36Palo Alto Networks security advisory (live-fetched)
Not affectedPanorama, Cloud NGFWNVD, vendor advisory (both explicitly state this)
CVSS (NVD)9.1 / CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:NNVD (independently confirmed via live NVD REST API query at report generation time)
CVSS (vendor's own rating)7.8 / HIGH, "Urgency HIGHEST"Palo Alto Networks security advisory — a real, unreconciled discrepancy with NVD's independently-computed 9.1/CRITICAL, documented here rather than silently resolved by picking one
EPSS score / percentile0.939 / 99.8th percentileFIRST.org EPSS API (live query at report generation time)
CISA KEV listedYES — added 2026-05-29, due date 2026-06-01 (passed)CISA KEV catalog (live-fetched)
Known ransomware campaign use (CISA's own flag)YES ("Known") — CISA does not name a specific ransomware family in this fieldCISA KEV catalog
Exploitation statusConfirmed — "limited exploit attempts on unpatched PAN-OS devices without mitigations applied"; vendor marks status "ATTACKED"Palo Alto Networks security advisory
Root causeRequires GlobalProtect portal or gateway configured with Authentication Override cookies enabled and a specific certificate configurationPalo Alto Networks security advisory
Workarounds(1) Use a dedicated certificate for Authentication Override cookies; (2) disable Authentication Override entirely in the portal/gateway configurationPalo Alto Networks security advisory
Publication / update datesPublished 2026-05-13; updated 2026-06-03 (FAQ section added)Palo Alto Networks security advisory

Technical Analysis

  1. Precondition — the target GlobalProtect portal or gateway has Authentication Override cookies enabled, together with a specific certificate configuration named but not further detailed in the vendor's public advisory. [Verified Fact] (HIGH CONFIDENCE)
  2. Exploitation — an attacker presents or manipulates the Authentication Override cookie mechanism such that the device trusts it without adequate validation (CWE-565: reliance on a cookie without validation and integrity checking), bypassing the authentication step the portal/gateway would otherwise require. [Verified Fact] (HIGH CONFIDENCE) — mechanism as described by the vendor; the exact byte-level or request-level exploitation detail has not been publicly disclosed by Palo Alto Networks, NVD, or CISA in sources available to this platform.
  3. Result — the attacker establishes an unauthorized VPN connection through the gateway, gaining network access equivalent to a legitimate remote-access VPN session. [Verified Fact] (HIGH CONFIDENCE)
  4. Scope boundary — Panorama and Cloud NGFW are explicitly not affected; this is scoped to PAN-OS GlobalProtect portal/gateway and the corresponding Prisma Access versions. [Verified Fact] (HIGH CONFIDENCE)
  5. Post-access activity — no source available to this platform (NVD, CISA, or the vendor advisory) describes what an attacker did after establishing unauthorized VPN access in the observed exploitation attempts. [Intelligence Gap]

MITRE ATT&CK Mapping

TechniqueNameTacticConfidenceEvidence
T1190Exploit Public-Facing ApplicationInitial AccessHIGHGlobalProtect portal/gateway is internet-facing infrastructure by design; the vulnerability is exploited directly against it, unauthenticated, per the vendor's own description
T1133External Remote ServicesInitial AccessMEDIUMThe resulting unauthorized access is, functionally, an external remote-access VPN session — the definitional example this ATT&CK technique describes — though the initial compromise mechanism is more precisely captured by T1190

No sub-technique specific to "authentication bypass via unvalidated cookie trust" was force-mapped. ATT&CK Enterprise does not have a clean, narrow technique for this exact mechanism the way it does for, say, credential dumping; T1556 (Modify Authentication Process) and T1550 (Use Alternate Authentication Material) were both considered and rejected — this platform's curated technique set does not currently include either, and forcing an uncurated ID onto a mechanism it does not precisely describe would repeat exactly the kind of imprecise mapping this platform's own quality gate exists to catch (see platform/open-issues.md Issue 3 and Issue 9 for two prior, real instances of this class of error found and fixed this session).

IOC Intelligence

No indicators of compromise (source IPs, target selection patterns, session artifacts, or any host/network indicator) have been publicly disclosed by NVD, CISA, or Palo Alto Networks as of this report. [Intelligence Gap] This is stated explicitly rather than populated with placeholder values. Organizations should prioritize the vendor's own patch/workaround guidance and GlobalProtect session review (below) over IOC blocklisting for this vulnerability, precisely because no blocklist-grade IOCs exist yet in public reporting available to this platform.

Behavioral Indicators & Detection Opportunities

Detection guidance (conceptual — PAN-OS log schema not independently verified)

No Sigma rule is provided for this report. The reason is stated explicitly rather than worked around with false precision: constructing a field-accurate Sigma rule requires the exact GlobalProtect/PAN-OS log field names for authentication-override-cookie events and VPN session establishment, and this platform did not independently confirm those field names against a live PAN-OS log export in the sources available for this report (NVD, CISA, and the vendor's own advisory text do not specify log schema). Publishing a Sigma rule with plausible-looking but unverified field names would present false technical precision — exactly the failure mode this platform's own quality gates and editorial standards exist to prevent (EIOS Layer 13; _gate_hype_language). The hunting guidance above describes the concept to search for; organizations should map it to their own confirmed PAN-OS/GlobalProtect log schema before building an automated rule.

Vulnerability Management Guidance

Patch prioritization: Emergency. KEV-listed status, a 99.8th-percentile EPSS score, and vendor-confirmed active exploitation together satisfy essentially any reasonable emergency-patch policy trigger, independent of which CVSS number (NVD's 9.1 or the vendor's own 7.8) an organization's policy keys off of. The federal remediation deadline (2026-06-01) is roughly eight weeks overdue as of this report's date — any organization still exposed should treat this as materially overdue, not merely "due soon."

If immediate patching is impossible: apply one of the vendor's two documented workarounds (dedicated certificate for Authentication Override cookies, or disabling Authentication Override) as an interim measure — these are vendor-published mitigations, not compensating controls this platform is inferring.

Regulatory Impact

US federal civilian executive branch agencies were required to remediate under the applicable CISA Binding Operational Directive by 2026-06-01 (already roughly eight weeks past as of this report's date). Non-federal organizations are not legally bound by that directive but should treat the KEV listing as an industry benchmark for reasonable patch timelines in any incident, audit, or regulatory inquiry.

Intelligence Gaps

Confidence Assessment

DimensionRatingRationale
Source ConfidenceHIGHNVD, CISA KEV catalog, and the vendor's own security advisory are primary/authoritative sources, independently queried live at report generation time
Evidence ConfidenceHIGHCore technical facts (CVSS, affected/fixed versions, KEV status, EPSS) independently confirmed via live queries against NVD, FIRST.org, CISA, and the vendor advisory
Technical ConfidenceMEDIUM-HIGHThe described mechanism and version ranges are vendor-stated and specific; the exact exploitation technique at the request/protocol level is not publicly detailed
Attribution ConfidenceLOWCISA confirms ransomware-campaign use generically; no specific actor or ransomware family is confirmed by an authoritative source
Detection ConfidenceLOWNo Sigma/detection artifact is provided in this report — deliberately, given unverified log-schema field names (see Detection Opportunities)
Overall ConfidenceHIGHGrounded in independently corroborated primary-source vulnerability data; appropriately hedged on the aspects (specific ransomware attribution, exact exploitation technique, IOCs) that are genuinely unconfirmed as of this report

References

  1. CVE-2026-0257 (live JSON record, queried at report generation time). NVD REST API. https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-0257
  2. CVE-2026-0257. National Vulnerability Database. https://nvd.nist.gov/vuln/detail/CVE-2026-0257
  3. EPSS score (live query at report generation time). FIRST.org. https://api.first.org/data/v1/epss?cve=CVE-2026-0257
  4. CISA Known Exploited Vulnerabilities Catalog (live feed, queried at report generation time). https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json
  5. Palo Alto Networks Security Advisory CVE-2026-0257 (live-fetched at report generation time). https://security.paloaltonetworks.com/CVE-2026-0257

Sentinel APEX Analyst Conclusion

CVE-2026-0257 is KEV-listed, EPSS-ranked in the top 0.2% of all scored vulnerabilities, vendor-confirmed under active exploitation, and roughly eight weeks past its federal remediation deadline as of this report — a combination of signals that independently justify emergency treatment without needing a specific ransomware family name attached to it. The platform's own aggregation pipeline surfaced a "Qilin ransomware" headline for this CVE with no locatable supporting source; this report deliberately does not repeat that claim as fact, and flags it instead as an unresolved reference pending verification, consistent with this platform's own reference-completeness standard (EIOS Layer 13; GIAAP v1). Enterprises running affected PAN-OS or Prisma Access builds should patch or apply one of the vendor's two documented workarounds immediately, and should review existing GlobalProtect sessions for anomalies predating remediation rather than treating a clean patch alone as proof no unauthorized access remains.


CyberDudeBivash® Sentinel APEX — Threat Intelligence Division