Automation Token Optimizer

• Outcome: Review
• Latest audit: Jun 20, 2026 at 5:53 PM UTC
• Version:
  1.0.0

Overview

The skill appears aimed at token or automation analysis, but it may read sensitive local session prompts without clear consent or redaction controls.

Install only if you are comfortable with the skill reading local automation/session history. Before using it, ask it to start with aggregate token metrics only, avoid displaying full prompts by default, redact sensitive task names or business details, and request confirmation before opening individual high-consumption prompt records.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• 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: 93% confidence
• Finding: The skill instructs the agent to read a local SQLite database containing automation session metadata and then inspect high-consumption task prompts, which may expose sensitive prompt contents, task names, schedules, and business context. Because the instructions do not require consent, minimization, or redaction before accessing and reporting this data, the skill increases the risk of unnecessary disclosure of local sensitive information.

VirusTotal

61/61 vendors flagged this skill as clean.