Threat reportVulnerabilityTL-2026-4823

Citrix NetScaler CVE-2026-3055 Memory Overread (Part 2) via /wsfed/passive?wctx — Session Leak, Exploited in the Wild

criticalACTIVE

Citrix NetScaler CVE-2026-3055 Memory Overread (Part 2) via (TL-2026-4823), also tracked as CitrixBleed 3, is a critical-severity software vulnerability scored CVSS 9.3, first published 2026-10-30. It has no confirmed attribution, affects Citrix (Cloud Software Group) NetScaler ADC and NetScaler Gateway, references 2 CVEs (CVE-2026-3055, CVE-2026-4368), maps to 6 MITRE ATT&CK techniques (T1078, T1190, T1212), and is covered by 9 detection rules and 7 indicators of compromise.

CVSS
9.3/10Critical
CVEs
2Referenced vulnerabilities
Techniques
6MITRE ATT&CK
Actors
0Not attributed
Detection rules
9SPL · KQL · Sigma
IOCs
7Indicators of compromise

Key facts for TL-2026-4823

Threat ID
TL-2026-4823
Also known as
CitrixBleed 3, NetScaler SAML IDP Memory Overread
Severity
CRITICAL
CVSS
9.3 (CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:L/SI:L/SA:L)
Status
ACTIVE
Category
VULNERABILITY
First published
Last reviewed
Attribution confidence
LOW
Motivation
UNKNOWN
Target sectors
government administration, finance, health, technology, enterprise
Target regions
Global
Detection rules
9
Indicators of compromise
7

Malware and tooling in Citrix NetScaler CVE-2026-3055 Memory Overread (Part 2) via

Malware and tooling: watchTowr detection artifact generator (Python)

How Citrix NetScaler CVE-2026-3055 Memory Overread (Part 2) via works

watchTowr Labs analyzes a second memory overread behind CVE-2026-3055 in Citrix NetScaler ADC/Gateway configured as a SAML IDP, triggered by sending /wsfed/passive?wctx with no value. Kilobytes of uninitialized memory leak base64-encoded in the NSC_TASS cookie, potentially exposing authenticated administrative session IDs; exploitation was observed from about 2026-03-27 and CISA added the CVE to KEV on 2026-03-30.

CVE-2026-3055 (Citrix advisory CTX696300, disclosed 2026-03-23, described as 'insufficient input validation leading to memory overread') affects NetScaler ADC and NetScaler Gateway only when the appliance is configured as a SAML Identity Provider (IDP). watchTowr Labs showed that the CVE actually covers at least two distinct out-of-bounds read (CWE-125) bugs on two endpoints: /saml/login (Part 1, malformed SAML request omitting the AssertionConsumerServiceURL attribute) and /wsfed/passive?wctx (Part 2).

In the Part 2 primitive, an unauthenticated attacker sends GET /wsfed/passive?wctx with the wctx parameter present but with no '=' sign and no value. The appliance checks only for the presence of the parameter before accessing the associated buffer, so it dereferences uninitialized memory. The vulnerable device answers with a 302 redirect and returns the leaked memory, base64-encoded, in the NSC_TASS cookie. watchTowr reports it works 'every(ish) time' and discloses kilobytes per request, with no data-quantity limit unlike the first primitive; repeated requests return different memory chunks, allowing an attacker to harvest session tokens, SAML assertions, LDAP credentials and active administrative session IDs for session hijacking. The pattern mirrors CitrixBleed 2 (CVE-2025-5777), and the issue has been nicknamed 'CitrixBleed 3' by some vendors.

Exploitation timeline: Defused Cyber and watchTowr's Attacker Eye honeypots saw reconnaissance against /cgi/GetAuthMethods (to fingerprint enabled authentication flows) on 2026-03-27, and watchTowr confirmed in-the-wild exploitation from known threat-actor source IPs by the same date. CrowdSec reported more than 40 probing IPs, mostly on AWS (AS16509). CISA added CVE-2026-3055 to the KEV catalog on 2026-03-30 with a federal deadline of 2026-04-02. A related, separate bug CVE-2026-4368 (race condition, CWE-362, CVSS v4 7.7) affects only build 14.1-66.54 and was not KEV-listed at the time of the Picus write-up.

Defenders should patch, confirm whether any appliance is configured as a SAML IDP, and, where exposure is possible, treat sessions and secrets as compromised: invalidate NSC_AAAC, NSC_TMAS and NSC_TASS sessions, force re-authentication, and rotate LDAP bind credentials, SAML signing keys and service-account passwords. No specific attacker IPs, user-agents or malware were published in the sources reviewed; actor attribution is unknown.

MITRE ATT&CK techniques used in TL-2026-4823

Initial Access

T1078 Valid Accounts; T1190 Exploit Public-Facing Application

Credential Access

T1212 Exploitation for Credential Access; T1539 Steal Web Session Cookie

lateral-movement

T1550.004 Web Session Cookie

Lateral Movement

T1550.004 Web Session Cookie

Reconnaissance

T1595.002 Vulnerability Scanning

Affected products and versions in Citrix NetScaler CVE-2026-3055 Memory Overread (Part 2) via

  • Citrix (Cloud Software Group) — NetScaler ADC and NetScaler Gateway (configured as SAML IDP)
    Vulnerable versions: 14.1 before 14.1-60.58; 14.1 before 14.1-66.59; 13.1 before 13.1-62.23; 13.1-FIPS/NDcPP before 13.1-37.262
    Fixed in: 14.1-60.58; 14.1-66.59; 13.1-62.23; 13.1-FIPS/NDcPP 13.1-37.262
  • Citrix (Cloud Software Group) — NetScaler Gateway/AAA vserver (CVE-2026-4368 only)
    Vulnerable versions: 14.1-66.54
    Fixed in: 14.1-66.59

Remediation for Citrix NetScaler CVE-2026-3055 Memory Overread (Part 2) via

Patches

  • NetScaler 14.1-60.58 or later
  • NetScaler 14.1-66.59 or later
  • NetScaler 13.1-62.23 or later
  • NetScaler 13.1-FIPS/NDcPP 13.1-37.262 or later (Citrix advisory CTX696300)

Immediate actions

  • Identify NetScaler ADC/Gateway appliances configured as a SAML IDP and prioritize them for patching
  • Invalidate NSC_AAAC, NSC_TMAS and NSC_TASS sessions and force re-authentication of all users
  • Rotate LDAP bind credentials, SAML signing keys and service-account passwords if exposure is possible
  • Hunt for GET /wsfed/passive?wctx (valueless) requests, /saml/login requests missing AssertionConsumerServiceURL, and external GET /cgi/GetAuthMethods

Workarounds

  • No workaround documented in the reviewed sources; exposure requires the appliance to be configured as a SAML IDP

Longer-term hardening

  • Reduce internet exposure of NetScaler management and authentication endpoints
  • Review whether the SAML IDP deployment model on the appliance is required
  • Monitor for anomalous VPN logins, ICA Proxy sessions outside business hours and RDP Proxy sessions from unfamiliar source IPs

CVEs associated with Citrix NetScaler CVE-2026-3055 Memory Overread (Part 2) via

CVE-2026-3055, CVE-2026-4368

Weaknesses (CWE) in Citrix NetScaler CVE-2026-3055 Memory Overread (Part 2) via

CWE-125, CWE-362

Timeline of Citrix NetScaler CVE-2026-3055 Memory Overread (Part 2) via

  • Citrix publishes advisory CTX696300 disclosing CVE-2026-3055 (insufficient input validation leading to memory overread) in NetScaler ADC/Gateway configured as a SAML IDP; found by Citrix's internal product security efforts
  • watchTowr confirms in-the-wild exploitation of the /wsfed/passive?wctx overread from known threat-actor source IPs; CrowdSec also dates early exploitation to this day
  • Defused Cyber and watchTowr Attacker Eye honeypots observe reconnaissance against /cgi/GetAuthMethods to fingerprint NetScaler authentication flows
  • watchTowr Labs publishes Part 2 analysis of the /wsfed/passive?wctx memory overread leaking data via the NSC_TASS cookie, with a detection artifact generator
  • CISA adds CVE-2026-3055 to the Known Exploited Vulnerabilities catalog
  • Federal remediation deadline for CVE-2026-3055 under CISA KEV

Sources cited for Citrix NetScaler CVE-2026-3055 Memory Overread (Part 2) via

Detection coverage for TL-2026-4823

As of 2026-10-30, Threadlinqs Intelligence publishes 9 detection rule(s) for TL-2026-4823 across Splunk SPL, Microsoft KQL and Sigma, covering 7 indicator(s) of compromise. The whole corpus is readable without an account; a free account unlocks full detection query text in Splunk SPL, Microsoft KQL and Sigma; paid tiers add raw indicator values, correlation and the MCP server. Threadlinqs MCP server · View plans.

9 detection rules (Splunk SPL, Microsoft KQL, Sigma) · Blue and above. Compare plans
7 indicators of compromise · Red and above. Compare plans

Threadlinqs Intelligence — Real-Time Threat Detection Platform

[ 0 threats ] [ 0 det ] [ CRIT: 0 ] [ HIGH: 0 ]
// threat_feed
$ sort --newest
Showing all threats

Live intelligence console

Threat level
Fig. 01 · Threat weatherIndexing the archive…
1 square = 1 threat · click to open

Latest Threats