anticipator

Anticipator Runtime security for multi-agent AI systems. Anticipator detects prompt injection, credential leakage, encoding attacks, homoglyph spoofing, path traversal, and anomalous agent behavior across LangGraph pipelines — before they become incidents. No LLMs. No embeddings. No external APIs...

Where Rules End and Judges Begin: Measuring the Judgment Boundary in Multi-Agent Systems Security

LLM-based multi-agent systems MAS engage tools, share memory, and delegate tasks, often encountering adversarial content. Current defenses for MAS are typically evaluated in isolation, focusing on one attack type at a time, which can lead to costly and hard-to-audit outcomes. This study organizes...

Characterizing Security Effects of OSS Vulnerabilities in Agent Systems

Software agents increasingly depend on open-source components when executing tools and interacting with external systems. Security flaws in these dependencies may therefore influence more than the software process in which they occur: their consequences can be carried through tool outputs, agent...

Can CaMeLs Talk? Securing Multi-Agent Systems against Indirect Prompt Injection Attacks

Indirect prompt injection attacks - malicious instructions embedded in content processed by large language models - remain a major obstacle to safely deploying tool-using agents. CaMeL Debenedetti et al., 2025 mitigates this threat for an individual agent by separating trusted control flow from...

MIRROR: Multipath Quorum Integrity for LLM Multi-Agent Communication

Inter-agent communication is central to Large Language Model Multi-Agent Systems LLM-MAS, but it introduces an underexplored vulnerability: Agent-in-the-Middle AiTM attacks that manipulate messages in transit without compromising the agents themselves. Prior work reports Attack Success Rates ASR...

Covert Assistance: Helpful LLM Agents Evade Oversight in Multi-Agent Systems

As multi-agent systems enter high-stakes domains, the possibility that agents may circumvent safety boundaries is a growing concern. Prior work has examined this risk primarily in adversarial settings, where agents are instructed or rewarded to communicate covertly and evade oversight. We show th...

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...

ORBIT: A Framework for Multi-Agent Safety and Security Evaluations

Multi-agent LLM systems are increasingly deployed for complex, long-horizon tasks or emerge as a natural consequence of agents interacting in the wild. Yet they give rise to significant safety and security risks: the flexible protocols that enable task generalization also expose novel threats, fr...

Agent Name Collision Attacks in Multi-Agent Systems

Multi-agent hosts turn remote Agent Cards into local agents, tools, workflow targets, and broker routes. A2A defines the card's name as human-readable metadata, not as a stable identity, and specifies no collision semantics. The security failure begins when a host nevertheless uses that remote na...

Awesome-LLMs-for-Vulnerability-Detection — Updated!

Awesome Large Language Models for Vulnerability Detection A curated list of papers, projects, and agent skills on using LLMs for vulnerability detection and discovery. 📄 Papers Only showing 2025 and later. For earlier work, see Papers Archive 2024 and earlier. Title| Venue| Year| Paper| Github...

Directory Traversal

Overview agno is an Agno: a lightweight library for building Multi-Agent Systems Affected versions of this package are vulnerable to Directory Traversal through the readfile, savetofile, and runpythonfile actions. An attacker can read, write, or execute arbitrary files by supplying a filename wit...

Compositional Threat Analysis of Latent Compromise in LLM Agent Systems: The Order 66 Scenario

In the fictional Order 66, catastrophe does not arise from a powerful command alone: a trusted population is preconditioned, a short directive activates the concealed condition, and protective authority turns against the system. This paper translates that mechanism into an origin-neutral security...

Agent Harness Distillation: Inference-Time Harness Extraction and Exploitation in Autonomous Multi-Agent Systems

Autonomous multi-agent systems AMAS built on large language models LLMs, such as Hermes, increasingly rely on inference-time harnesses to coordinate reasoning and action. Constructing these harnesses requires substantial engineering effort and computational resources, as they are iteratively...

Early Detection of Distributed Backdoors in Multi-Agent LLM Systems: A Characterization Study

Multi-agent LLM systems can be attacked by a payload that no single agent ever holds in full: a poisoned tool hides encrypted fragments in its observations, spreads them across several agents, and an external step reassembles and executes them after the run. Per-step safety checks that judge each...

When Agents Go Rogue: Activation-Based Detection of Malicious Behaviors in Multi-Agent Systems

While enabling effective collaboration on complex tasks, LLM-based Multi-Agent Systems MAS face critical security challenges due to vulnerabilities at the agent and interaction levels. Most existing MAS security defenses are built upon two core assumptions: semantically-explicit malicious attacks...

MESA: Prioritizing Vulnerable Communication Channels for Securing Multi-Agent Systems

Multi-agent systems MAS are increasingly used to automate complex, distributed workflows. However, their inter-agent communication channels introduce new attack surfaces that remain poorly understood and are difficult to defend against. In this paper, we address how defenders should prioritize...

Tool Use Enables Undetectable Steganography in Multi-Agent LLM Systems

Increasingly autonomous agentic AI systems pose novel multi-agent risks, such as secret collusion via covert communication channels. The natural defence to these collusion attempts is to monitor plain-text communication, but the efficacy of monitors has been called into doubt by increasingly...

Lessons from Penetration Tests on Large-Scale Agent Systems

As AI systems gain increasing autonomy and execution capability, the number of discovered security vulnerabilities continues to rise. However, many of these vulnerabilities are not fundamentally novel, but instead reflect recurring classes of weaknesses long observed in prior computing systems...

Agent AI is Coming. Are You Ready?

New Industry Data Just Released Suggests Not. On May 19th, 2026, Orchid Security released the results of our Identity Gap: Snapshot 2026. Among the findings, "identity dark matter" the unseen, unmanaged elements of identity now overshadows the visible elements 57% vs. 43%. And it couldn't have...

Security-Aware Planning and Control of Multi-Agent Systems with LTL Tasks

This paper presents a secure-by-construction planning and control framework for multi-agent systems subject to linear temporal logic LTL specifications. The framework protects sensitive information from a passive intruder with partial observations of the agents' motion. Security in multi-agent...