Naruto Multi-Agent

• Outcome: Pass
• Latest audit: May 28, 2026 at 10:48 AM UTC
• Version:
  1.0.0

Overview

This skill is a disclosed Naruto-themed dispatcher that delegates user tasks to named sub-agents, with no install hooks, executable code, hidden data flow, or credential handling found.

Install this only if you want ordinary work requests routed into persistent named sub-agent sessions. Avoid including secrets or sensitive production context unless you are comfortable with that context being copied into a delegated task, and give clear limits before asking it to perform account-changing or destructive work.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Vague Triggers

Medium

• Confidence: 92% confidence
• Finding: The skill uses very broad activation language such as responding whenever it 'receive[s] a mission request,' but it does not clearly define what qualifies as a mission versus normal conversation, analysis, or harmless roleplay. In a dispatcher skill that automatically delegates via `sessions_spawn`, ambiguous triggering can cause unintended sub-agent execution on ordinary user inputs, increasing the risk of prompt injection propagation, unnecessary task spawning, and loss of user intent control.

VirusTotal

64/64 vendors flagged this skill as clean.