AgentDoG 1.5: A Lightweight and Scalable Alignment Framework for AI Agent Safety and Security

Modern open-world agents such as OpenClaw exhibit powerful cross-environment execution capabilities yet introduce broad new safety risk sources. Meanwhile, advanced frontier AI models drastically lower attack barriers, rendering current agent alignment frameworks inadequate for real-world...

GHSA-WXXX-GVQV-XP7P LiteLLM has a sandbox escape in custom-code guardrail

Impact The POST /guardrails/testcustomcode endpoint runs user-supplied Python inside a hand-rolled sandbox. The sandbox can be escaped using bytecode-level techniques, allowing arbitrary code execution in the proxy process — which runs as root in the default Docker image. Reaching the endpoint...

LiteLLM has a sandbox escape in custom-code guardrail

Impact The POST /guardrails/testcustomcode endpoint runs user-supplied Python inside a hand-rolled sandbox. The sandbox can be escaped using bytecode-level techniques, allowing arbitrary code execution in the proxy process — which runs as root in the default Docker image. Reaching the endpoint...

Duplicate Advisory: LiteLLM has a sandbox escape in custom-code guardrail

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-wxxx-gvqv-xp7p. This link is maintained to preserve external references. Original Description LiteLLM through 2026-04-08 allows remote attackers to execute arbitrary code via bytecode rewriting at the...

GHSA-3926-2JVF-FG29 Duplicate Advisory: LiteLLM has a sandbox escape in custom-code guardrail

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-wxxx-gvqv-xp7p. This link is maintained to preserve external references. Original Description LiteLLM through 2026-04-08 allows remote attackers to execute arbitrary code via bytecode rewriting at the...

We Found Eight Attack Vectors Inside AWS Bedrock. Here's What Attackers Can Do with Them

AWS Bedrock is Amazon's platform for building AI-powered applications. It gives developers access to foundation models and the tools to connect those models directly to enterprise data and systems. That connectivity is what makes it powerful – but it’s also what makes Bedrock a target. When an AI...

ChainFuzzer: Greybox Fuzzing for Workflow-Level Multi-Tool Vulnerabilities in LLM Agents

Tool-augmented LLM agents increasingly rely on multi-step, multi-tool workflows to complete real tasks. This design expands the attack surface, because data produced by one tool can be persisted and later reused as input to another tool, enabling exploitable source-to-sink dataflows that only...

Improper Neutralization of Input Used for LLM Prompting

Overview @n8n/n8n-nodes-langchain is a Affected versions of this package are vulnerable to Improper Neutralization of Input Used for LLM Prompting via the Guardrail node. An attacker can modify workflow input to circumvent intended restrictions by crafting specific input values. Workaround This...

n8n has a Guardrail Node Bypass

Impact An end user interacting with a workflow that uses the Guardrail node could craft an input that bypasses the default guardrail instructions. Patches The issue has been fixed in n8n version 2.10.0. Users should upgrade to this version or later to remediate the vulnerability. Workarounds If...

GHSA-FVFV-PPW4-7H2W n8n has a Guardrail Node Bypass

Impact An end user interacting with a workflow that uses the Guardrail node could craft an input that bypasses the default guardrail instructions. Patches The issue has been fixed in n8n version 2.10.0. Users should upgrade to this version or later to remediate the vulnerability. Workarounds If...

Grok apologizes for creating image of young girls in “sexualized attire”

Another AI system designed to be powerful and engaging ends up illustrating how guardrails routinely fail when development speed and feature races outrun safety controls. In a post on X, AI chatbot Grok confirmed that it generated an image of young girls in “sexualized attire.” The potential...

Eurostar Accused Researchers of Blackmail for Reporting AI Chatbot Flaws

Researchers discovered critical flaws in Eurostar’s AI chatbot including prompt injection, HTML injection, guardrail bypass, and unverified chat IDs - Eurostar later accused them of blackmail...

Black-Box Guardrail Reverse-Engineering Attack

Large language models LLMs increasingly employ guardrails to enforce ethical, legal, and application-specific constraints on their outputs. While effective at mitigating harmful responses, these guardrails introduce a new class of vulnerabilities by exposing observable decision patterns. In this...

BreakFun: Jailbreaking LLMs Via Schema Exploitation

The proficiency of Large Language Models LLMs in processing structured data and adhering to syntactic rules is a capability that drives their widespread adoption but also makes them paradoxically vulnerable. In this paper, we investigate this vulnerability through BreakFun, a jailbreak methodolog...

Active Honeypot Guardrail System: Probing and Confirming Multi-Turn LLM Jailbreaks

Large language models LLMs are increasingly vulnerable to multi-turn jailbreak attacks, where adversaries iteratively elicit harmful behaviors that bypass single-turn safety filters. Existing defenses predominantly rely on passive rejection, which either fails against adaptive attackers or overly...

What happened in Vegas (that you actually want to know about)

Welcome to this week's edition of the Threat Source newsletter. Last week I flew 5,000 miles to Las Vegas for Black Hat USA. After navigating the casino carpet labyrinth and finding the only venue in Nevada that serves a proper English breakfast tea with milk lifesaver, I've decided Black Hat fee...

Exploit for Incorrect Authorization in Sudo_Project Sudo

sudo CVE-2025 Toolkit Unified scanner, benign proof-of-...

OneShield -- the Next Generation of LLM Guardrails

The rise of Large Language Models has created a general excitement about the great potential for a myriad of applications. While LLMs offer many possibilities, questions about safety, privacy, and ethics have emerged, and all the key actors are working to address these issues with protective...

PolyGuard: Massive Multi-Domain Safety Policy-Grounded Guardrail Dataset

Whitepaper called PolyGuard: Massive Multi-Domain Safety Policy-Grounded Guardrail Dataset...

LlamaFirewall: an Open Source Guardrail System for Building Secure AI Agents

Large language models LLMs have evolved from simple chatbots into autonomous agents capable of performing complex tasks such as editing production code, orchestrating workflows, and taking higher-stakes actions based on untrusted inputs like webpages and emails. These capabilities introduce new...