ClawHub
Back to skill

Security audit

deepen

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed research-note skill that writes a local, reusable knowledge base, with privacy considerations but no hidden or destructive behavior found.

Install this if you want an agent to create and reuse local research notes over time. Set the knowledge-base location deliberately, avoid pointing it at public repos or mixed personal/work/private notes, and do not store secrets, credentials, or case-specific confidential information in reusable domain-reference notes.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Ssd 3

Medium
Confidence
91% confidence
Finding
The skill explicitly instructs the agent to build a compounding, cross-session knowledge base optimized for future model loading. That creates durable retention of user-provided and researched content beyond the immediate task, which can unintentionally preserve sensitive data, copyrighted material, or confidential notes and make them available to later runs.

Ssd 3

Medium
Confidence
94% confidence
Finding
The skill directs the agent to resolve a filesystem path, write notes there by default, and even return notes inline for manual saving when persistence is unavailable. This creates a natural-language exfiltration and retention channel where sensitive user or system-derived content could be written to disk or surfaced in chat output without strong consent, sanitization, or boundary checks.

Ssd 3

Medium
Confidence
88% confidence
Finding
The instruction to search the entire knowledge-base root for related topics by keyword and meaning encourages broad retrieval across existing notes, including semantically similar content. In practice, this can pull unrelated sensitive material into the current task context and cause cross-project data leakage or unauthorized reuse of notes the current user did not intend to access.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.