ResearchResearch paperSafety & Alignment · Large Language Models · Agents & Tool Use1 source · Oct 13, 2026

SkillScreen: High-Recall Screening of Malicious Agent Skills with Decision Models

We present SkillScreen, a compact decision model trained to score skill maliciousness directly from SKILL.md in a single forward pass.

Agent skills turn natural-language instructions into actions, creating a security challenge: legitimate automation and malicious behaviour can look deceptively similar.
A practical screen must catch nearly every attack while keeping the review load manageable.
On MaliciousSkillBench, SkillScreen flags 20-25% of benign skills at 99% recall on unseen attack families, substantially fewer than the word-level classifier and prompted LLM judges tested.
Our results demonstrate the value of task-trained decision models for high-recall skill screening.

[1]SkillScreen: High-Recall Screening of Malicious Agent Skills with Decision Models
arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 13, 05:03 AM
We present SkillScreen, a compact decision model trained to score skill maliciousness directly from SKILL.md in a single forward pass.
Agent skills turn natural-language instructions into actions, creating a security challenge: legitimate automation and malicious behaviour can look deceptively similar.

Extractive summary: sentences quoted from the sources.

Before this