ClawHub
Back to skill

Security audit

Deslop

Security checks across malware telemetry and agentic risk

Overview

This is a markdown-only writing helper that rewrites user-provided text to sound less AI-generated, with one routing-scope caveat but no hidden access or execution behavior.

Reasonable to install. Use it when you explicitly want de-AI-tone rewriting, and review outputs carefully because the skill intentionally changes style and its broad triggers may activate on some generic writing-improvement requests.

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 (1)

Vague Triggers

Medium
Confidence
92% confidence
Finding
Several triggers are broad enough to activate on ordinary writing requests, which can cause the wrong skill to run and silently rewrite user content under a specialized 'de-AI-tone' policy. In an agent system, overbroad routing increases the chance of unintended prompt takeover at the skill-selection layer, especially for normal requests like style edits or human-writing help.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal