ClawHub
Back to skill

Security audit

Rust工具包-免费版

Security checks across malware telemetry and agentic risk

Overview

This Rust helper is not clearly malicious, but it asks for shell and file-writing power while also claiming to be knowledge-only, so it should be reviewed before installation.

Install only if you are comfortable with an agent potentially running shell commands and modifying files during Rust tasks. Prefer a version that removes automatic ping behavior or clearly limits command execution and file writes to user-approved Rust project operations.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • 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
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
94% confidence
Finding
The skill is presented as a Rust best-practices knowledge assistant, but it requests powerful `exec` and `write` capabilities and later describes command-based network diagnostics. This creates a scope mismatch that can cause an agent to perform local command execution unrelated to the user’s Rust question, increasing the risk of unintended system actions and trust abuse.

Context-Inappropriate Capability

Medium
Confidence
96% confidence
Finding
Automatically running `ping` for retry and network diagnostics is not justified for a Rust coding guidance skill and introduces unnecessary command execution and network interaction. Even though `ping` is not the most dangerous command, it can leak environmental information, bypass user expectations, and normalize autonomous shell/network activity in a skill that claims to be knowledge-focused.

Intent-Code Divergence

Medium
Confidence
92% confidence
Finding
The documentation claims the skill is a pure Markdown, no-external-API tool, yet elsewhere it authorizes command execution and discusses network-dependent behavior. This inconsistency is dangerous because it can mislead users and hosting agents about the real operational risk, resulting in overtrust and execution of actions that would not be expected from a documentation-only skill.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill declares `exec`/`write` capabilities and describes automatic network diagnostic behavior without a clear user-facing warning or consent model. In context, that is more dangerous because the skill markets itself as a free Rust practice assistant, so users are likely to assume low operational risk and may not expect filesystem or shell side effects.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.