ClawHub
Back to skill

Security audit

python-debug-assistant

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only Python debugging helper with no executable code, credential access, persistence, or hidden data flow shown in the artifacts.

Reasonable to install for Python debugging. Users should avoid pasting secrets, private tokens, or proprietary code into debugging prompts unless they are comfortable sharing that content with the agent, and non-Chinese users may need to ask the assistant to respond in their preferred language.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger phrases are broad enough to match many ordinary Python help requests, which can cause the skill to activate when the user did not specifically intend to invoke this workflow. Unintended activation is a policy and routing risk because it may override a more appropriate general assistant behavior or apply rigid output formatting and assumptions to unrelated requests.

Natural-Language Policy Violations

Medium
Confidence
79% confidence
Finding
The skill description and instructions enforce Chinese output without offering user choice, which can cause unsafe or incorrect handling when the user's language, environment, or expectations differ. While not a classic security flaw, it is a robustness and trust issue that can degrade comprehension of debugging guidance and increase the chance of user error when remediation steps are misunderstood.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.