Multi Persona Ai

Security checks across malware telemetry and agentic risk

Overview

This skill is a small local voting tool for generating persona settings, with no evidence of hidden data access, networking, or automatic system changes.

Install this only if you want a local helper for persona voting. Run it in a dedicated project directory so .consensus_votes.json is easy to manage, and review any generated persona rules before using them to configure an agent.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (2)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger phrase "consensus persona engine" is broad and provides no contextual constraints for when the skill should activate. In an agent ecosystem, underspecified triggers can cause accidental invocation in unrelated conversations, leading the skill to influence persona, autonomy, or safety settings when it was not explicitly requested.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The reset command irreversibly deletes the local vote file without confirmation, safeguards, or backup. While limited to a predictable local file, this can still cause unintended data loss if invoked accidentally or by a higher-level agent/tooling flow without user review.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal