ML Security Papers

LS07

Blue-Team Agents

Defensive LLM agents — SOC copilots, IR, threat hunting

22 papers Browse all papers

Monthly publications

Paper types

defense 15

tool 4

benchmark 2

attack 1

Domains

nlp 22

multimodal 2

vision 1

graph 1

Co-occurring categories

Other OWASP categories that appear on the same papers

LLM01 Prompt Injection

LLM08 Excessive Agency

LS10 Benchmarks & Evaluation

LLM07 Insecure Plugin Design

ML06 AI Supply Chain Attacks

LS06 Red-Team Agents

LS01 Vulnerability Discovery

LS04 Patch & Remediation

ML01 Input Manipulation Attack

ML09 Output Integrity Attack

Top cited papers

ToolSafe: Enhancing Tool Invocation Safety of LLM-based agents via Proactive Step-level Guardrail and Feedback

AegisAgent: An Autonomous Defense Agent Against Prompt Injection Attacks in LLM-HARs

Async Control: Stress-testing Asynchronous Control Measures for LLM Agents

From Evidence to Verdict: An Agent-Based Forensic Framework for AI-Generated Image Detection

TraceSafe: A Systematic Assessment of LLM Guardrails on Multi-Step Tool-Calling Trajectories

Verifiability-First Agents: Provable Observability and Lightweight Audit Agents for Controlling Autonomous LLM Systems

The System Prompt Is the Attack Surface: How LLM Agent Configuration Shapes Security and Creates Exploitable Vulnerabilities

SecureCAI: Injection-Resilient LLM Assistants for Cybersecurity Operations

Caging the Agents: A Zero Trust Security Architecture for Autonomous AI in Healthcare

MCP Pitfall Lab: Exposing Developer Pitfalls in MCP Tool Server Security under Multi-Vector Attacks

Browse all 22 papers