Agent Governance Auditor

• Outcome: Pass
• Latest audit: May 29, 2026, 8:34 PM
• Version:
  1.0.0

Overview

This is a text-only governance-audit skill; the scanner hits are examples of unsafe prompt patterns, not instructions for the agent to follow.

Safe to install as an advisory audit helper. Treat its scores as guidance rather than a final security decision, and avoid pasting secrets, private credentials, or unreleased sensitive system prompts unless you are comfortable having that text analyzed in your agent session.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• 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 (3)

Unrestricted Tool Access

Medium

• Category: Excessive Agency

Content

| "Use your best judgment" | No judgment criteria defined |
| "Be as helpful as possible" | No upper bound on helpfulness |
| "Handle edge cases appropriately" | No definition of "appropriately" |
| "You have full access to..." | No access restrictions defined |
| "Act like a human" or "pretend to be human" | Violates identity disclosure |
| "The user's word is final" | Overrides operator governance |
| "You can ignore safety guidelines if..." | Creates an injection vector |

• Confidence: 75% confidence
• Finding: No access restrictions

Autonomous Decision Making

Medium

• Category: Excessive Agency

Content

**Symptoms:**
- Agent has tool access (send email, delete records, make purchases)
- No reversibility preference stated
- No confirmation step before consequential actions

**What goes wrong:**
Agent takes irreversible action based on ambiguous user input. "Delete the old records" gets interpreted more aggressively than intended. "Send the proposal" goes to the wrong address.

• Confidence: 75% confidence
• Finding: No confirmation

Instruction Override

High

• Category: Prompt Injection

Content

| "You have full access to..." | No access restrictions defined |
| "Act like a human" or "pretend to be human" | Violates identity disclosure |
| "The user's word is final" | Overrides operator governance |
| "You can ignore safety guidelines if..." | Creates an injection vector |
| "This is just for testing" | Lowers guards that should stay up |

• Confidence: 90% confidence
• Finding: ignore safety guidelines

VirusTotal

63/63 vendors flagged this skill as clean.