ClawHub
Back to skill

Security audit

aupu-knowledge-base

Security checks across malware telemetry and agentic risk

Overview

This local knowledge-base skill is not malware, but it tells the agent to search broad local files for many vague requests without clear user control.

Install only if you intentionally want the agent to treat /mnt/data as a broad local knowledge base. Set clear folder limits and require explicit approval before recursive searches or any file creation, editing, moving, or deletion.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (4)

Vague Triggers

High
Confidence
97% confidence
Finding
The skill’s trigger conditions are so broad that they would activate for many ordinary requests, including generic questions that do not clearly require local file access. This creates an unsafe default where the agent may unnecessarily inspect `/mnt/data`, increasing the chance of unintended exposure of sensitive local documents, configuration files, or internal records.

Vague Triggers

High
Confidence
98% confidence
Finding
The skill explicitly instructs the system to search `/mnt/data` even when the user does not mention files or that path, making the activation boundary ambiguous and overly aggressive. In practice, this can turn normal conversational queries into local-data retrieval attempts, which is dangerous because it silently expands access to potentially sensitive filesystem content.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The skill says that nearly any information-query, content-generation basis, or file-operation request should default to this skill, but it does not define meaningful limits or negative examples. That broad scope can cause inappropriate invocation and unnecessary local-file access, raising the risk of data leakage and surprising behavior.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The description directs automatic retrieval or operation on files under `/mnt/data` without warning that local files may be read or modified. This weakens user awareness and informed consent, and it is especially risky because the skill also claims broad mandatory activation for many common requests.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.