Psyke SEO & GEO Audit

• Outcome: Review
• Latest audit: May 29, 2026 at 9:28 PM UTC
• Version:
  1.0.0

Overview

This skill performs a legitimate SEO audit workflow, but it can publish client audit material publicly without a clear consent step.

Install only if you are comfortable with a workflow that may crawl client sites, query third-party AI/search tools, and publish audit results. Before use, require explicit approval for any GitHub or Vercel deployment, prefer private or local output for client work, and review the deck for confidential business details before sharing.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• 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

Findings (3)

Context-Inappropriate Capability

Medium

• Confidence: 92% confidence
• Finding: The skill goes beyond generating an audit deck and instructs the agent to publish the output to a public GitHub repository and a production Vercel deployment. That creates an unnecessary data exposure path: audit content can include client names, URLs, competitor analysis, traffic estimates, and other potentially sensitive business information, all without an explicit consent gate or visibility check.

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: The trigger phrases are broad enough to activate on ordinary requests like 'run an audit' or 'audit this site', which can cause the skill to run in contexts the user did not clearly intend. In combination with the skill's crawl, analysis, and deployment steps, overbroad activation increases the chance of unintended network access, data collection, and public publication.

Missing User Warnings

High

• Confidence: 97% confidence
• Finding: The skill explicitly directs deployment to a public GitHub repo and a production Vercel site, but does not warn the user that the output will be publicly accessible. Because the deck may contain client-specific findings, competitor benchmarks, and AI visibility results, this can leak sensitive commercial information immediately and at scale.

VirusTotal

65/65 vendors flagged this skill as clean.