CVE-2025-33245

NVIDIA NeMo Framework contains a vulnerability where malicious data could cause remote code execution. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, information disclosure, and data tampering...

CVE-2025-33245

NVIDIA NeMo Framework contains a vulnerability where malicious data could cause remote code execution. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, information disclosure, and data tampering...

CVE-2025-33243

NVIDIA NeMo Framework contains a vulnerability where an attacker could cause remote code execution in distributed environments. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, information disclosure, and data tampering...

CVE-2025-33243

NVIDIA NeMo Framework contains a vulnerability where an attacker could cause remote code execution in distributed environments. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, information disclosure, and data tampering...

CVE-2025-33243

Summary (CVE-2025-33243) NVIDIA NeMo Framework is affected; a vulnerability could allow remote code execution in distributed environments. The issue impacts NVIDIA NeMo Framework across all platforms and versions prior to 2.6.1. The security bulletin lists the update to version 2.6.1 or later as ...

CVE-2025-33243

NVIDIA NeMo Framework contains a vulnerability where an attacker could cause remote code execution in distributed environments. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, information disclosure, and data tampering...

CVE-2025-33241

NVIDIA NeMo Framework contains a vulnerability where an attacker could cause remote code execution by loading a maliciously crafted file. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, information disclosure, and data tampering...

CVE-2025-33241

NVIDIA NeMo Framework contains a vulnerability where an attacker could cause remote code execution by loading a maliciously crafted file. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, information disclosure, and data tampering...

CVE-2025-33241

CVE-2025-33241 affects NVIDIA NeMo Framework. A remote-code-execution vulnerability can be triggered by loading a maliciously crafted file, with potential impacts including code execution, privilege escalation, information disclosure, and data tampering. NVIDIA’s security bulletin (Feb 2026) prov...

CVE-2025-33236

NVIDIA NeMo Framework contains a vulnerability where malicious data created by an attacker could cause code injection. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, information disclosure, and data tampering...

CVE-2025-33236

NVIDIA NeMo Framework contains a vulnerability where malicious data created by an attacker could cause code injection. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, information disclosure, and data tampering...

CVE-2025-33236

CVE-2025-33236 is associated with the NVIDIA NeMo Framework. The Red Hat, CIRCL, NVD, OSV, and NVIDIA bulletins corroborate a vulnerability where attacker‑crafted data can cause code injection, potentially leading to code execution, privilege escalation, information disclosure, and data tampering...

CVE-2025-33236

NVIDIA NeMo Framework contains a vulnerability where malicious data created by an attacker could cause code injection. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, information disclosure, and data tampering...

“Good enough” emulation: Fuzzing a single thread to uncover vulnerabilities

A Cisco Talos researcher worked around the limitations of hardware-level Code Read-out Protection RDP on the Socomec DIRIS M-70 gateway by pivoting from physical debugging to a "good enough" emulation approach. By focusing on emulating only the single thread responsible for Modbus protocol handli...

PT-2026-20399

Name of the Vulnerable Software and Affected Versions NVIDIA NeMo Framework affected versions not specified Description The NVIDIA NeMo Framework has a flaw that allows an attacker to create malicious data that could lead to code injection. Exploitation of this issue may result in code execution,...

PT-2026-20410

Name of the Vulnerable Software and Affected Versions NVIDIA NeMo Framework affected versions not specified Description The NVIDIA NeMo Framework has a flaw that could allow a remote attacker to execute code. This is achieved by tricking a user into opening a specially designed file. Exploitation...

CVE-2025-33245: Deserialization of Untrusted Data

NVIDIA NeMo Framework contains a vulnerability where malicious data could cause remote code execution. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, information disclosure, and data tampering...

CVE-2025-33246: Improper Neutralization of Special Elements used in a Command

NVIDIA NeMo Framework for all platforms contains a vulnerability in the ASR Evaluator utility, where a user could cause a command injection by supplying crafted input to a configuration parameter. A successful exploit of this vulnerability might lead to code execution, escalation of privileges,...

Recursive Language Models for Jailbreak Detection: A Procedural Defense for Tool-Augmented Agents

Jailbreak prompts are a practical and evolving threat to large language models LLMs, particularly in agentic systems that execute tools over untrusted content. Many attacks exploit long-context hiding, semantic camouflage, and lightweight obfuscations that can evade single-pass guardrails. We...

Automating Agent Hijacking Via Structural Template Injection

Agent hijacking, highlighted by OWASP as a critical threat to the Large Language Model LLM ecosystem, enables adversaries to manipulate execution by injecting malicious instructions into retrieved content. Most existing attacks rely on manually crafted, semantics-driven prompt manipulation, which...