de-ai-flavor

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:42 PM UTC
• Version:
  1.0.0

Overview

The skill has a disclosed file-editing workflow that should be used carefully, but the supplied evidence does not show hidden, deceptive, or unrelated behavior.

Before installing, use this skill only on files you can restore from version control or backups. Ask the agent to show a proposed patch before writing when the file is important, and review the final diff carefully.

SkillSpector

By NVIDIA

Vulnerability Patterns

• 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
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The skill explicitly says it will directly modify the user-supplied file and only show a diff afterward, which means a write occurs before clear confirmation. In an agent setting, this creates an unsafe file-modification pattern: a user may expect analysis and suggestions, but the skill performs destructive changes immediately, increasing the risk of unintended edits or corruption of important content.

VirusTotal

66/66 vendors flagged this skill as clean.