ClawHub
Back to skill

Security audit

vaDeepresearch

Security checks across malware telemetry and agentic risk

Overview

This is a transparent research-report skill that may over-activate on broad analysis prompts but does not show hidden, destructive, persistent, or credential-seeking behavior.

Install this if you want structured research reports with citations and Markdown files. Be aware it may activate for broad 'analyze' or 'compare' requests, so specify when you only want a short answer or do not want browsing/file generation.

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 (2)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The skill description and trigger guidance use very broad phrases such as deep analysis, help me analyze, and similar common requests, which can cause the skill to activate for routine conversations far outside narrowly scoped research tasks. Over-broad activation increases the chance of unintended tool use, unnecessary file generation, and context takeover by this skill when a simpler or safer response path was appropriate.

Vague Triggers

Medium
Confidence
87% confidence
Finding
The trigger section defines ambiguous activation criteria in both Chinese and English without strong scope boundaries, while also encouraging use for broad comparison and analysis requests. In an agent setting, this can lead to over-activation, causing the model to perform unnecessary browsing, source collection, and mandatory markdown artifact generation on prompts that only loosely resemble research tasks.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.