Benchmark RadarAI BENCHMARK PROFILE
MalSkillBench
MalSkillBench is a runtime-verified benchmark of malicious agent skills, with 3,944 malicious skills labeled along a taxonomy, measuring detection tool effectiveness.
- Released
- 2026-06-05
- Readiness
- Paper only
- Primary field
- Transport & Logistics
Why it matters
Evaluates detection tools for hybrid code-prompt skills, potentially informing supply chain security, but lacks public artifacts for reuse.
Motivation
AI coding agents such as Claude Code and Gemini CLI increasingly extend themselves with third-party skills: markdown packages bundling natural-language instructions, executable scripts, and tool permissions.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.