ClawHub
Back to skill

Security audit

Pinch to Post - Manage WordPress sites through WP Pinch MCP server

Security checks across malware telemetry and agentic risk

Overview

This is a clearly disclosed WordPress management skill, with real administrative power that appears purpose-aligned and controlled through the WP Pinch MCP server.

Install this only for WordPress sites you intend to manage from an agent. Use a least-privileged WordPress Application Password or the OpenClaw Agent role, keep read-only mode or write budgets enabled where practical, review WP Pinch audit logs, and be aware that broad triggers may make the skill available in loosely related conversations.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger list includes generic terms such as "post," "publish," and especially "site management," which can cause the skill to activate in conversations that are not clearly intended to manage WordPress. In a high-impact skill that can modify content, plugins, themes, users, and orders, unintended invocation increases the risk of accidental administrative actions or disclosure of site data.

Overly Broad Trigger

Low
Category
Trigger Abuse
Confidence
94% confidence
Finding
The trigger `wp` is extremely short and likely to collide with unrelated text, abbreviations, or incidental mentions. Because this skill exposes powerful WordPress management capabilities, accidental invocation can create unnecessary exposure to administrative actions, even if downstream tools still enforce permissions.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.