Agent Performance Profiler

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:41 PM UTC
• Version:
  1.0.0

Overview

This is an instruction-only agent performance profiling skill, with no executable installer or credential request, but its automatic trigger wording is broader than ideal.

Install only if you want an agent-performance profiling prompt. Prefer manual invocation or narrower trigger settings, and avoid running it over sensitive conversations or tool traces until you know what data the host agent exposes to skills and whether profiling reports are stored.

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: 93% confidence
• Finding: The trigger set is broad enough to activate on ordinary discussion of performance, optimization, tokens, or latency, which can cause unintended skill invocation. In an agent environment, overbroad activation can alter routing, inject irrelevant instructions into unrelated conversations, and increase the chance of prompt-surface abuse or operational confusion.

VirusTotal

66/66 vendors flagged this skill as clean.