Nm Conserve Agent Expenditure

• Outcome: Pass
• Latest audit: Jun 7, 2026 at 9:44 PM UTC
• Version:
  1.0.3

Overview

This appears to be an informational helper for reviewing multi-agent efficiency, with only a minor concern that its activation wording is broad.

This should be reasonable to install if you want guidance around multi-agent efficiency reviews. Be aware it may trigger more often than a narrowly named skill, so verify that it is relevant when it appears in adjacent workflow or efficiency conversations.

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: 95% confidence
• Finding: The manifest includes a free-form natural-language trigger, "evaluating parallel agent efficiency or after a multi-agent run," which is broader and less predictable than specific keyword-style triggers. This can cause the skill to activate in loosely related contexts, increasing the chance of unintended invocation and prompt-surface expansion, though the content of this skill is informational rather than directly privileged or code-executing.

VirusTotal

53/53 vendors flagged this skill as clean.