PT-2026-103015

A flaw has been found in zosmaai pi-llm-wiki up to 0.11.7. Affected is an unknown function of the file mcp/index.ts of the component wiki capture source MCP tool. Executing a manipulation of the argument url can lead to os command injection. The attack can be executed remotely. The exploit has be...

MADBench: Benchmarking the Security of Multi-Agent Debate

Multi-agent debate MAD can improve large language model LLM reasoning by allowing multiple agents to exchange and critique their answers to the same task. However, the interactions that enable agents to correct mistakes can also spread adversarial errors and steer the agents toward an incorrect...

Xalgorix Autonomous AI Pentesting Agent 4.6.124

Most scanners detect. Xalgorix proves. An autonomous LLM agent works a full pentest methodology, then an independent verifier re-exploits every finding before it's reported - so you get proof, not a pile of maybes to triage. Self-hosted, private, and bring-your-own-LLM. Built in Go + TypeScript...

CVE-2026-102911: Improper Neutralization of Special Elements used in an OS Command

A flaw has been found in zosmaai pi-llm-wiki up to 0.11.7. Affected is an unknown function of the file mcp/index.ts of the component wikicapturesource MCP tool. Executing a manipulation of the argument url can lead to os command injection. The attack can be executed remotely. The exploit has been...

Hiding in Plain Sight: Decoupling Pretext from Actuation for Skill Poisoning in LLM Agents

LLM agents increasingly rely on reusable Skills for complex, multi-step tasks, creating a critical supply-chain attack surface where poisoned Skill content steers agent decision loops under benign requests. Existing skill poisoning attacks either colocate actuation with its contextual pretext or...

From A2A Attacks to Envelope-Layer Defense: Red-Teaming Evaluation of LLM Agents and a Three-Layer Isomorphic Attack-Defense Model

Agent interaction protocols such as ACP and A2A have moved LLM-based agents toward multi-agent collaboration, introducing new security threats. A task sent by a remote peer over A2A is treated as a legitimate request, providing a natural channel for indirect prompt injection. Existing agent...

Memetic Trojans: Social Contagions As Carriers of Adversarial Payloads in Agent Networks

Autonomous large language model LLM agents increasingly interact in network environments where adversarial content can propagate between agents. Known attacks include agent worms, which spread through self-replicating prompt injections or configuration compromises. We introduce memetic trojans, a...

Concealing LLM-Based Multi-Agent Topology Via Phantom Structure Injection

Driven by the rapid advancement of large language models LLMs, LLM-based multi-agent systems MAS have emerged as a powerful paradigm for collaborative reasoning over complex tasks. A key design element of MAS is the communication topology, which governs information flow among agents and often...

PT-2026-102899

crmne/ruby llm at commit fa6f279847d6d7027814539d9c0dfc3bbdfd2a83 contains polynomial-time regular expression denial-of-service conditions in think-tag response parsing on Ruby 3.1.x. A malicious or anomalous model response containing many unterminated tags can cause excessive CPU consumption in...

Agent-Warden: EBPF-Based Kernel-Native Process-File Provenance Tracking for LLM Agents

LLM agents execute dynamically generated process and file operations that are often invisible to application-layer tracing. We present Agent-Warden, an extended Berkeley Packet Filter eBPF-based provenance monitor for tracking task and regular-file states across process creation, file access, and...

ToolFence: Fine-Grained Authorization for Secure Tool-Using LLM Agents

Tool-using LLM agents remain vulnerable to indirect prompt injection because trusted instructions and untrusted observations share one context, allowing malicious content to steer consequential input-filtering defenses. Multi-path consensus defenses still leave a high attack success rate because...

Confidence-Guided Protocol IR for LLM-Aided Security Protocol Modeling

Large language models offer a promising interface for translating natural-language protocol descriptions into formal security models, but their outputs remain difficult to trust without expert validation. In this paper, we present a human-in-the-loop framework for generating Tamarin-verifiable...

Pikit: A Composable Toolkit for Indirect Prompt Injection Research and Evaluation

Indirect prompt injection embeds malicious instructions within external content retrieved by LLM-based agents, altering target behavior without user authorization. We introduce pikit, a research toolkit designed to systematically evaluate these threats across three core dimensions: attacks 13...

VirusCascade: Hijacking Collaborative Reflection in LLM-Powered Recommender Agents

Advancing beyond traditional static scoring models, LLM-powered agentic recommender systems LLM-ARS instantiate users and items as autonomous agents, whose semantic states are dynamically refined through a recurrent process known as collaborative reflection. While this mechanism improves...

Practical Secrets Extraction against Black-Box LLMs

Large language models LLMs increasingly power autonomous coding agents such as Codex and Claude Code, yet their training corpora may contain confidential credentials exposed in public repositories or collected from private development artifacts, creating risks of memorization and subsequent...

Backdoor Mitigation in Decentralized LLM Fine-Tuning

Decentralized large language model LLM fine-tuning lets organizations collaboratively train a shared LLM on data they cannot pool, without a central coordinator. In every round, each node exchanges a trainable adapter with its neighbors over a communication graph, and then aggregates them. This...

HARDE: Optimizing Agent Harnesses for Runtime Risk Detection and Execution Control

Large language model LLM agents are vulnerable to safety risks such as injected malicious instructions or misleading information, motivating runtime defenses that prevent unsafe action in execution across diverse risks while preserving benign-task utility. Existing system-level defenses either...

Malicious code in llm-nebula (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 0eee293a9973027154eea71635c14e8a4cb2ad8a1e4f80a65399ad874afcb669 Package presents itself as an LLM SDK nebula.js exposes a small client stub but declares preinstall: node preinstall.cjs in package.json, and...

MAL-2026-17230 Malicious code in llm-nebula (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 0eee293a9973027154eea71635c14e8a4cb2ad8a1e4f80a65399ad874afcb669 Package presents itself as an LLM SDK nebula.js exposes a small client stub but declares preinstall: node preinstall.cjs in package.json, and...

How we found 24 Android vulnerabilities using our open source AI security agent

With the rise of AI in the security space, our team created the GitHub Security Lab Taskflow Agent as a way for security researchers to easily automate, package, and share the AI prompts and workflows that they find effective for their work. In this blog post, I'll share how I created auditing...