No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →
by protectskills · Claude Skill · ★ 72
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
🔒 Is MaliciousAgentSkillsBench safe to install? View the security audit →
"Do Not Mention This to the User": Detecting and Understanding Malicious Agent Skills This repository contains a comprehensive security benchmark dataset and evaluation framework for Claude Code Agent Skills. The paper reports a three-tiered, nested dataset of 98,380 skills from two major platforms (skills.rest and skillsmp.com): 4,287 statically-flagged suspicious candidates (Tier 2), of which 157 are behaviorally-confirmed malicious skills (Tier 3). The 157 confirmed skills are a verified subset of the 4,287 candidates — not a separate group — and the candidates are themselves a subset of the 98,380-skill snapshot.
| Stars | 72 |
| Forks | 6 |
| Language | Python |
| Category | Claude Skill |
| License | MIT |
| Quality Score | 60.1371893702931/100 |
| Last Updated | 2026-08-03 |
| Created | 2026-02-07 |
| Platforms | claude-code, python |
| Est. Tokens | ~17k |
Looking for a MaliciousAgentSkillsBench alternative? If you're comparing MaliciousAgentSkillsBench with other claude skill tools, these 1 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
A tool for autonomous agents like OpenClaw to discover and learn skills autonomously
Explore other popular claude skill tools:
MaliciousAgentSkillsBench is A Security Benchmark for Claude Code Agent Skills. It is categorized as a Claude Skill with 72 GitHub stars.
MaliciousAgentSkillsBench is primarily written in Python.
You can find installation instructions and usage details in the MaliciousAgentSkillsBench GitHub repository at github.com/protectskills/MaliciousAgentSkillsBench. The project has 72 stars and 6 forks, indicating an active community.
MaliciousAgentSkillsBench is released under the MIT license, making it free to use and modify according to the license terms.
The top alternatives to MaliciousAgentSkillsBench on Agent Skills Hub include skyll. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: