ClawHub
Back to skill

Security audit

Deslop the Copy

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed text-rewriting skill with no code execution or data access, but it may rewrite more aggressively than some users expect.

Install this if you want a strong 'make it sound less AI-written' editor. Review outputs carefully for changed tone, lost formality, added opinions, or altered authorship cues, especially for academic, legal, medical, financial, workplace, or brand-sensitive writing.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (3)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The trigger phrases are broad enough to catch ordinary editing requests such as 'clean up this copy' or 'fix it,' which can cause the skill to activate outside its intended scope. In this skill's context, that matters because it steers the model toward specifically making text appear less AI-generated, a capability that can be misused to evade AI-origin scrutiny or override a user's neutral editing intent.

Vague Triggers

High
Confidence
92% confidence
Finding
The trigger description is broad enough to match common rewriting requests like 'clean up,' 'fix,' or 'make this sound more natural,' which can cause the skill to activate when the user did not specifically ask for AI-style humanization. That can lead to unintended rewriting, tone changes, or policy-relevant transformations being applied too aggressively, especially in contexts where preserving register, formality, or authorship cues matters.

Natural-Language Policy Violations

Medium
Confidence
87% confidence
Finding
The instruction to 'use contractions everywhere' and to contract forms a native speaker would naturally contract imposes a specific dialect and fluency norm without user consent. This can overwrite user-preferred voice, create locale bias, and produce inappropriate edits for formal, non-native, accessibility, legal, or brand-sensitive writing contexts.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.