ClawHub
Back to skill

Security audit

Junyi Vault Organizer

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent Obsidian note organizer that writes local text files to a configured vault, with the main risk being accidental persistent saves from broad trigger phrases.

Install only if you want the agent to create and update real files in your Obsidian vault. Configure vault_path narrowly, keep backups before first use, and avoid using vague save/remember phrases around sensitive content unless you intend it to be archived.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill clearly instructs reading and writing physical files in a user-configured Obsidian vault, yet no explicit permissions are declared. That creates a trust and policy gap: a user or platform may not realize the skill can persist or modify disk content, increasing the chance of unintended data exposure or unauthorized file changes if the trigger fires unexpectedly.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger phrases are extremely broad, including common requests like 'save this', 'remember this', and general mentions of storing information. In a conversational assistant, this can cause accidental activation and persistent storage of sensitive material the user did not intend to archive, especially because the skill writes to disk rather than ephemeral memory.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
Although the body later mentions physical-file storage, the triggerable skill behavior does not prominently and consistently warn users at activation time that their content will be written to persistent files on local disk. This matters because users may assume 'remember' means transient assistant memory, not durable storage in a searchable vault, leading to inadvertent retention of personal or confidential information.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.