What is the CVSS score of CVE-2026-25481?

CVE-2026-25481 has a CVSS 3.1 base score of 9.6 (Critical). CVSS vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H. EPSS exploitation probability: 0.1%.

Which versions of AI / ML are affected by CVE-2026-25481?

Organizations using Langroid versions prior to 0.59.32 are at critical risk due to a bypass vulnerability that circumvents previous security fixes, potentially allowing attackers to execute arbitrary…. A patch is available.

Is there a public PoC exploit available for CVE-2026-25481?

Yes, a public proof-of-concept exploit is available for CVE-2026-25481. Prioritise patching immediately. EPSS: 0.1%.

How to fix or mitigate CVE-2026-25481?

Within 24 hours: Inventory all systems running Langroid and identify affected versions; notify development and operations teams of critical status. Within 7 days: Apply vendor patch to upgrade Langroid to version 0.59.32 or later across all development, staging, and production environments; conduct verification testing. Within 30 days: Conduct security audit of all Langroid-dependent applications and review access logs for potential exploitation attempts during the vulnerability window.

AI / ML CVE-2026-25481

CRITICAL

Code Injection (CWE-94)

2026-02-04 security-advisories@github.com

GHSA-x34r-63hx-w57f

Code Injection AI / ML Langroid

9.6

CVSS 3.1 · GitHub Advisory

Severity by source

GitHub Advisory PRIMARY

9.6 CRITICAL

AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H

Primary rating from GitHub Advisory · only source for this CVE.

CVSS VectorGitHub Advisory

Attack Vector

Network

Attack Complexity

Low

Privileges Required

None

User Interaction

Required

Scope

Changed

Confidentiality

High

Integrity

High

Availability

High

Lifecycle Timeline

Analysis Generated

Mar 12, 2026 - 21:54 vuln.today

PoC Detected

Feb 20, 2026 - 21:20 vuln.today

Public exploit code

Patch released

Feb 20, 2026 - 21:20 nvd

Patch available

CVE Published

Feb 04, 2026 - 20:16 nvd

CRITICAL 9.6

DescriptionGitHub Advisory

Langroid is a framework for building large-language-model-powered applications. Prior to version 0.59.32, there is a bypass to the fix for CVE-2025-46724. TableChatAgent can call pandas_eval tool to evaluate the expression. There is a WAF in langroid/utils/pandas_utils.py introduced to block code injection CVE-2025-46724. However it can be bypassed due to _literal_ok() returning False instead of raising UnsafeCommandError on invalid input, combined with unrestricted access to dangerous dunder attributes (__init__, __globals__, __builtins__). This allows chaining whitelisted DataFrame methods to leak the eval builtin and execute arbitrary code. This issue has been patched in version 0.59.32.

AnalysisAI

Langroid LLM framework prior to 0.5 has a code injection vulnerability (CVSS 9.6) allowing attackers to execute arbitrary code through the AI agent system.

Technical ContextAI

Langroid < 0.5 has a CWE-94 code injection vulnerability that allows attackers to inject and execute arbitrary code through the LLM agent interaction system.