51mee Interview Evaluation Report
Security checks across malware telemetry and agentic risk
Overview
This is a simple interview-report prompt skill with no executable code, credentials, persistence, or external-service behavior evident in the artifacts.
Safe to install for drafting interview evaluation reports. Review the skill’s privacy claim against your OpenClaw runtime before using real candidate data, and keep final hiring decisions under human review.
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.