Security audit
persona-evaluator
Security checks across malware telemetry and agentic risk
Overview
The skill is a disclosed persona-auditing workflow; its local reads, OpenPersona CLI use, remote-agent review steps, and optional persona fixes fit its stated purpose.
Install only if you are comfortable letting an agent run OpenPersona CLI commands against persona packs and, in black-box mode, interact with or review public material from remote agents. Review targets before allowing remediation commands such as refine --apply, update, or direct persona.json edits.
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
63/63 vendors flagged this skill as clean.
Static analysis
No suspicious patterns detected.