Security audit

Db Explorer

Security checks across malware telemetry and agentic risk

Overview

This database helper is mostly coherent, but it includes high-impact restore and migration commands without enough specific safeguards.

Install only if you are comfortable with an agent using database client commands against systems you specify. Use read-only or least-privileged credentials, avoid production/admin access unless explicitly needed, review every command before execution, and require separate confirmation before any restore, import, migration, full export, or write operation.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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

Findings (5)

Description-Behavior Mismatch

Medium

Confidence: 95% confidence
Finding: The skill is presented as a database exploration and debugging tool, but it also includes restore and migration capabilities that can directly alter or overwrite data. That scope expansion is dangerous because an agent may invoke destructive or high-impact operations under the umbrella of a seemingly read-oriented skill, increasing the chance of unauthorized modification or operational misuse.

Intent-Code Divergence

Medium

Confidence: 96% confidence
Finding: The safety section promises read-only-by-default behavior and confirmation before writes, but later commands include restore and import actions without equivalent safeguards. This inconsistency can mislead downstream agents or users into trusting the skill as safe while still exposing direct write paths that can modify or destroy database contents.

Context-Inappropriate Capability

Low

Confidence: 82% confidence
Finding: Including backup capability verification as a routine post-connection step expands the operational scope from inspection into privileged administrative actions. Even if not immediately destructive, normalizing this behavior can prompt unnecessary access to sensitive data or execution of commands that exceed the user's stated intent and least-privilege expectations.

Vague Triggers

Medium

Confidence: 87% confidence
Finding: The activation cues are broad enough to match ordinary conversation about databases, which increases the likelihood of accidental invocation. Because the skill includes sensitive database access and some modifying/admin operations, over-triggering materially raises the chance of exposing data, running unintended queries, or steering the agent into risky workflows without clear user intent.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: Backup, restore, and migration instructions are high-risk operations that can overwrite data, import untrusted content, or move sensitive records across systems, yet the documentation does not present explicit overwrite/data-loss warnings. In an agent setting, omission of these warnings makes it more likely that such commands are treated as routine and executed without adequate safeguards.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.