Auto Router

• Outcome: Review
• Latest audit: May 28, 2026 at 11:53 AM UTC
• Version:
  1.0.0

Overview

This skill appears intended to auto-route user prompts to different AI models, but its routing scope, external data/cost implications, and local script dependency are not clearly bounded enough for automatic approval.

Review before installing. Only use this if you understand which models/providers receive your prompts, what billing limits apply, and what the missing local auto-router.js does. Avoid routing secrets, private customer data, credentials, or regulated data until the skill documents explicit consent, provider, cost, and exclusion rules.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• 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

Findings (2)

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: The phrase indicating the skill 'automatically analyzes your messages and routes to the optimal model' is overly broad and does not define scope, consent, or trigger conditions. In this skill's context, that can cause users to unknowingly have arbitrary prompts processed or forwarded, including to external paid models, which creates privacy, cost, and policy risks.

Missing User Warnings

High

• Confidence: 95% confidence
• Finding: The skill presents automatic model routing and cost savings but does not clearly warn that some user prompts may be sent to paid third-party API models such as Claude Sonnet or Opus. That omission is dangerous because users may disclose sensitive information under the assumption processing is local, leading to unexpected external data transfer, billing, and compliance exposure.

VirusTotal

66/66 vendors flagged this skill as clean.