Agent Tool Marketplace
Security checks across malware telemetry and agentic risk
Overview
The skill artifacts are operational and high-authority, but their sensitive actions are clearly disclosed, scoped, and gated by user or staff confirmation.
Install this only for a ClawHub/Convex maintainer context where the agent is expected to use repo credentials, GitHub/Convex tooling, and staff moderation authority. Before running moderation, production migration, or autoreview helper commands, confirm the target, reason, deployment, and whether full-access nested review is acceptable; use the no-yolo option when you do not want nested review to bypass sandbox prompts.
SkillSpector
By NVIDIA
Vulnerability Patterns
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
VirusTotal
64/64 vendors flagged this skill as clean.