WordPress Optimizer

• Outcome: Review
• Latest audit: May 29, 2026 at 9:27 PM UTC
• Version:
  1.0.0

Overview

This skill has a legitimate WordPress optimization purpose, but it asks for powerful site credentials and can make broad live-site changes without enough scoping, rollback, or credential-handling detail.

Install only if you trust the provider and can tolerate changes to the target WordPress site. Prefer testing on staging first, take a backup, use a temporary scoped account or revocable token instead of a primary admin password, ask for a clear list of changes before applying them, and revoke credentials after use.

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 (4)

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The skill describes endpoints that can automatically optimize and harden a live WordPress site but does not warn users that these actions may change production configuration, plugins, themes, or security settings. In an agent setting, omission of this warning increases the chance of unintended destructive or service-impacting changes being triggered against live systems without informed consent or staging review.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill explicitly solicits WordPress admin credentials and tokens in request bodies without any warning about secret handling, storage, redaction, or transmission risk. In an agent ecosystem, users may paste highly privileged credentials that could be exposed in logs, telemetry, chat history, or to an untrusted third-party API, leading to full site compromise.

Vague Triggers

Medium

• Confidence: 86% confidence
• Finding: The API exposes broad operations like /optimize-wp and /tune-wp with minimal scoping, no documented exclusions, and no indication of what concrete actions will be performed on the target WordPress instance. In a tool that accepts site credentials or tokens, this ambiguity is risky because an agent could trigger sensitive configuration changes, plugin/theme modifications, or security-affecting actions beyond a user's intended scope.

Natural-Language Policy Violations

Low

• Confidence: 80% confidence
• Finding: The schema requires a raw password for WPSiteRequest without documenting why direct credential handling is necessary, what consent model is expected, or how the secret will be protected. In this skill context, collecting administrator credentials for a third-party optimization tool increases the risk of credential theft, over-privileged access, insecure storage, and accidental reuse across sites.

VirusTotal

65/65 vendors flagged this skill as clean.