Assistant Reliability Watchtower

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:37 PM UTC
• Version:
  0.1.0

Overview

The available evidence shows a purpose-aligned skill with a least-privilege documentation gap, not hidden or malicious behavior.

Before installing, confirm you are comfortable with the skill running its local Python/shell helper and reading the intended workspace or environment-derived paths. Prefer running it in a limited project directory and review any wrapper commands before allowing them to act on sensitive files or secrets.

SkillSpector

By NVIDIA

Vulnerability Patterns

• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• 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)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 93% confidence
• Finding: The skill instructs operators to run a Python wrapper and references environment variables, filesystem reads, and shell commands, but it declares no corresponding permissions. This creates a capability/permission mismatch that can undermine security review and least-privilege enforcement, because users or orchestration layers may not realize the skill can access repo files, environment-derived paths, or execute subprocesses.

VirusTotal

67/67 vendors flagged this skill as clean.