ClawHub
Back to skill

Security audit

Humanize Text

Security checks across malware telemetry and agentic risk

Overview

This skill edits text to remove AI-style writing patterns, and its file access is disclosed and aligned with that purpose.

Install only if you want a skill that may edit files you ask it to process. Use it on copies or version-controlled documents for important content, avoid applying it to code, configs, or sensitive records unless intended, and review diffs before keeping changes.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding
The skill instructs the agent to read a file, transform its contents, and write it back, but the metadata does not declare corresponding permissions or constraints. Undeclared write capability is risky because it can surprise the user, bypass expected consent boundaries, and lead to unintended modification of local content.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger phrases are generic writing requests such as 'make this sound human' and 'clean up AI writing,' which can overlap with normal conversation and cause the skill to activate when the user did not intend file or transformation behavior. Unintended invocation becomes more concerning here because the skill includes file-processing and write-back instructions.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly says to read the file, apply transformations, and write it back, but it does not require a prominent warning or explicit approval before modifying user data. This creates a real integrity risk: a mistaken invocation or overbroad transformation could silently alter documents, including important technical or policy content.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.