MaxKB
CVE-2025-66446
HIGH
Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessAuthenticate as low-privileged MaxKB user
-
ExploitWin race to overwrite dynamic linker
-
ExecutionPrivileged process loads tampered file
-
ImpactEscalate privileges toward root
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A06:2025 - Insecure Design
ASVS 5.0
V15 - Secure Coding and Architecture Concurrency control, locking
MITRE ATT&CK
1 technique across 1 tactic: Privilege Escalation The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up. |
| Exploitation Conditions | Requires an authenticated low-privileged MaxKB account (PR:L) plus successfully winning a race condition (AC:H, CWE-362) against improperly-permissioned critical files - specifically the built-in dynamic linker (ld.so) and other system files that MaxKB leaves writable. Exploitation is limited to MaxKB versions 2.3.1 and below; version 2.4.0 remediates it. Limiting factors: the attacker must already hold valid low-privilege access, must reliably hit the narrow race-timing window, and a higher-privileged process must subsequently load the overwritten file for escalation to occur. No user interaction is required (UI:N). |
| Impact | Signals point to a moderate, not urgent, priority despite the 7.5 'High' CVSS. The NVD vector (AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H) requires an already-authenticated low-privileged account (PR:L) and a high-complexity race-condition window (AC:H, consistent with the CWE-362 classification), which meaningfully constrains real-world exploitation. EPSS reinforces this: 0.35% probability at the 26th percentile indicates low near-term mass-exploitation expectation. There is no CISA KEV listing and no public proof-of-concept in the provided intelligence, so exploitation status is 'no public exploit identified at time of analysis' - this is not confirmed active exploitation. One tension to verify: the vector is AV:N, but overwriting the on-disk dynamic linker and critical files is inherently a local filesystem action, so network reachability likely depends on authenticated MaxKB application features being abused - verify the true attack surface with the vendor advisory (GHSA-5xx2-3q9w-jpgf). Also note a scope judgment call: NVD rates S:U, yet escalation from the application context to the host/root arguably crosses a trust boundary (S:C); this is debatable and does not change the overall moderate rating. Net: patch on a normal prioritized cycle, not as an emergency. |
| Exploit ScenarioDerived | A low-privileged authenticated MaxKB user abuses improper file permissions to target the built-in dynamic linker or other critical files, winning a race condition (CWE-362) to overwrite one with attacker-controlled content. When a higher-privileged process subsequently loads the tampered file, the attacker's code executes with elevated (potentially root) privileges. No public exploit was identified at time of analysis, and the high attack complexity (AC:H) reflects the timing precision needed to win the race. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Oct 07, 2026
Privilege escalation in MaxKB 2.3.1 and earlier allows an authenticated, low-privileged user to overwrite the built-in dynamic linker and other critical system files, ultimately enabling code execution in the context of a higher-privileged process. Exploitation requires winning a narrow race condition and a privileged process subsequently loading the overwritten file, which is why the base 7.5 'High' score (AV:N/AC:H/PR:L/UI:N) reflects high impact but non-trivial attack conditions. No public exploit code or confirmed active exploitation was identified at time of analysis, and EPSS is low at 0.35% (26th percentile), so this should be treated as a moderate, not urgent, patch-priority item; version 2.4.0 remediates the flaw.
RemediationAI
Vendor-released patch: upgrade to MaxKB 2.4.0 (https://github.com/1Panel-dev/MaxKB/releases/tag/v2.4.0), which explicitly fixes the file-overwrite permission bypass for CVE-2025-66446 and the related concurrent-condition issue CVE-2025-66419. If immediate upgrade is not possible, treat the flaw as an authenticated abuse path and reduce the attack surface: remove or disable unnecessary low-privileged MaxKB accounts so that only trusted operators can reach the vulnerable functionality, and audit and tighten filesystem permissions on the container/host so that ld.so and other system files are not writable by the MaxKB service user - the trade-off is that over-restrictive permissions can break legitimate MaxKB features or container startup and must be validated in staging. Running MaxKB with a read-only root filesystem and a non-root service user, where the application supports it, removes the ability to overwrite critical system files in the first place, at the cost of requiring writable volumes to be explicitly mounted for application data; note that these are compensating controls, not equivalents to the 2.4.0 patch, and patch status is confirmed by the vendor advisory (https://github.com/1Panel-dev/MaxKB/security/advisories/GHSA-5xx2-3q9w-jpgf).
Affects your stack
Find out whether the stacks and servers you run are on an affected version.
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
MaxKB is an open-source enterprise AI assistant from 1Panel-dev, commonly deployed as a containerized web application. The root cause is classified as CWE-362 (Concurrent Execution Using Shared Resource with Improper Synchronization), combined here with improper file permissions on security-critical paths: the built-in dynamic linker (ld.so) and other system files are left writable by the MaxKB service. An authenticated low-privileged user (PR:L in the assessed CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H vector) can race a write against those improperly permissioned files, effectively injecting code into the loader or other shared resources; a higher-privileged process that later loads the tampered artifact then executes attacker-controlled content, yielding privilege escalation. The high AC:H rating reflects the difficulty of reliably hitting the race-timing window, and UI:N confirms no victim interaction is needed. The same 2.4.0 release also addressed a related concurrency-based permission bypass tracked as CVE-2025-66419, indicating a cluster of file-permission and race-condition weaknesses in how MaxKB handles system-level files.
Know when Maxkb gets its next vulnerability
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H
Authenticated low-priv access (PR:L) and a timing race (AC:H, CWE-362) are required; AV:N retained for app-mediated reach; overwriting the linker yields full host compromise (C/I/A:H).
Primary rating from NVD.
CVSS VectorNVD
DescriptionNVD
MaxKB is an open-source AI assistant for enterprise. Versions 2.3.1 and below have improper file permissions which allow attackers to overwrite the built-in dynamic linker and other critical files, potentially resulting in privilege escalation. This issue is fixed in version 2.4.0.
Executive ViewAI
MaxKB versions 2.3.1 and earlier contain a high-severity privilege escalation flaw that allows an authenticated low-privileged user to overwrite critical system files and potentially execute code as a higher-privileged process. While exploitation requires a narrow race condition and no public exploit or active attacks have been identified, a successful compromise could lead to full host takeover, data exposure, and lateral movement within the environment. Leadership should prioritize upgrading to version 2.4.0 as a moderate-risk remediation, given the low likelihood but high potential impact.
Apply vendor patch.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
Privilege escalation occurs when an attacker leverages flaws in access control mechanisms to gain permissions beyond what they were originally granted.
Race conditions occur when multiple concurrent requests exploit the brief time gap between checking a condition and acting upon it (TOCTOU - Time of Check, Time of Use).
Affected Products (CPE)NVD
MaxKB prior to version 1.10.8-lts contains an incomplete sandbox implementation that only blacklists binary execution in
Command injection in MaxKB before 2.10.5-lts allows unauthenticated remote attackers to execute arbitrary shell commands
Sandbox escape and privilege escalation in MaxKB 2.3.1 and earlier allow an attacker who can submit or trigger custom Py
Server-side request forgery in MaxKB before 2.10.0 allows any authenticated user holding the default workspace USER role
Authenticated workspace members on MaxKB deployments prior to 2.10.5-lts can escalate to root-level command execution by
Server-side request forgery in MaxKB (open-source AI assistant from 1Panel-dev) allows an authenticated user with permis
OS command injection in MaxKB (1Panel-dev's open-source enterprise AI assistant) before 2.10.0-lts lets an authenticated
Authentication bypass in MaxKB (1Panel-dev) versions prior to 2.9.0 allows remote unauthenticated attackers to invoke we
Cross-workspace read and modification of model parameter defaults in MaxKB 2.10.3-lts and earlier is possible for an aut
Stored Cross-Site Scripting (XSS) via Eval Injection in MaxKB's Markdown rendering engine allows authenticated users to
Stored Cross-Site Scripting (XSS) in MaxKB 2.7.1 and below allows authenticated users to inject malicious JavaScript thr
MaxKB, an open-source enterprise AI assistant by 1Panel-dev, stores user passwords as unsalted MD5 hashes, exposing all
NVD
CVE.org
EPSS
GitHub
Exploit-DB