collaborators

8 papers

cs.CR2026

VIGIL: Runtime Enforcement of Behavioral Specifications in AI Agent Skills

Ying Li, Yanju Chen, Hongbo Wen +5

Agentic systems increasingly act through third-party skills, allowing model-generated decisions to affect files, communication channels, and cyber-physical devices. These skills of…

cs.CR2026

Hunting Vulnerability Variants in AI Infra: Measurement and Reference-Driven Detection

Tian Dong, Yanjun Chen, Shoufeng Zhang +6

AI infra has become a shared execution layer for model training, deployment, and agent orchestration. Because many projects reimplement similar model-centric workflows, a vulnerabi…

cs.CR2026

No Attack Required: Semantic Fuzzing for Specification Violations in Agent Skills

Ying Li, Hongbo Wen, Yanju Chen +3

LLM-powered agents can silently delete documents, leak credentials, or transfer funds on a routine user request, not because the agent was attacked, but because the skill it invoke…

cs.CR2026

Options, Not Clicks: Lattice Refinement for Consent-Driven MCP Authorization

Ying Li, Yanju Chen, Peiran Wang +4

As Model Context Protocol adoption grows, securing tool invocations via meaningful user consent has become a critical challenge, as existing methods, broad always allow toggles or…

cs.CR2026

Semia: Auditing Agent Skills via Constraint-Guided Representation Synthesis

Hongbo Wen, Ying Li, Hanzhi Liu +4

An agent skill is a configuration package that equips an LLM-driven agent with a concrete capability, such as reading email, executing shell commands, or signing blockchain transac…

cs.CR2026

Synthesizing Multi-Agent Harnesses for Vulnerability Discovery

Hanzhi Liu, Chaofan Shou, Xiaonan Liu +4

LLM agents have begun to find real security vulnerabilities that human auditors and automated fuzzers missed for decades, in source-available targets where the analyst can build an…