ML Security Papers

LLM07

Insecure Plugin Design

Vulnerabilities in LLM plugins, function calling, tool use

111 papers Browse all papers

Monthly publications

Paper types

attack 30

defense 29

benchmark 28

survey 20

tool 4

Domains

nlp 111

multimodal 9

reinforcement-learning 2

graph 1

vision 1

audio 1

Co-occurring categories

Other OWASP categories that appear on the same papers

LLM01 Prompt Injection

LLM08 Excessive Agency

ML06 AI Supply Chain Attacks

LLM06 Sensitive Information Disclosure

LLM04 Model Denial of Service

ML09 Output Integrity Attack

ML10 Model Poisoning

ML02 Data Poisoning Attack

ML01 Input Manipulation Attack

Top cited papers

Securing the Model Context Protocol (MCP): Risks, Controls, and Governance

MCPGuard : Automatically Detecting Vulnerabilities in MCP Servers

Systematization of Knowledge: Security and Safety in the Model Context Protocol Ecosystem

Agent Skills Enable a New Class of Realistic and Trivially Simple Prompt Injections

Agentic AI Security: Threats, Defenses, Evaluation, and Open Challenges

When MCP Servers Attack: Taxonomy, Feasibility, and Mitigation

A Survey on Agentic Security: Applications, Threats and Defenses

Agent Skills in the Wild: An Empirical Study of Security Vulnerabilities at Scale

Securing AI Agent Execution

ToolTweak: An Attack on Tool Selection in LLM-based Agents

Browse all 111 papers