Perspectium
Security checks across malware telemetry and agentic risk
Overview
The skill set is coherent and purpose-aligned, with sensitive actions disclosed and generally gated by user intent or existing auth controls.
Before installing, understand that these skills are intended for maintainers and can guide agents to use authenticated CLIs for moderation, GitHub comments, proof publishing, deployment, and nested code review. Use them only in a trusted ClawHub checkout with credentials you intend the agent to use, and prefer the documented confirmation and opt-out flags for high-impact actions.
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.