ClawHub
Back to skill

Security audit

Plain Writing

Security checks across malware telemetry and agentic risk

Overview

This is a writing-help skill with local linting tools; the file and script use is disclosed and aligned with editing drafts.

Install if you are comfortable with a writing skill that may run bundled local Python checks against drafts you provide. Avoid passing highly sensitive documents unless you trust the local environment, and confirm file paths before allowing lint or before/after comparison commands.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill declares no permissions, but its instructions direct the agent to read local reference files, write temporary drafts, and execute local Python scripts. That creates undeclared file and shell capabilities, which can bypass user expectations and platform policy checks and may expose local data or trigger unintended command execution in an agent environment.

Tp4

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding
The published description presents the skill as a prose-editing aid, but the body instructs the agent to run local linters, compare files, expand inputs into local paths, and perform package self-checks. This mismatch is dangerous because users may invoke what appears to be a harmless writing skill without realizing it can access local files and execute shell commands, increasing the risk of unexpected data exposure or tool use.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.