Security audit

Db Connector Free

Security checks across malware telemetry and agentic risk

Overview

This database guidance skill is mostly advisory, but it requests command execution and file write authority without clear enough scoping or user controls.

Review this skill before installing. It does not show malicious behavior or hidden code, but you should treat it as capable of running local commands and writing files. Use it only if you are comfortable with live database/network diagnostics, and do not provide production credentials unless the agent session has explicit user-controlled safeguards.

SkillSpector

By NVIDIA

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Behavioral ASTexec() Call, eval() Call, Dynamic Import
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (6)

Intent-Code Divergence

Medium

Confidence: 94% confidence
Finding: The skill claims to be a pure knowledge-base with no extra dependencies, yet the manifest grants exec/write and the document instructs environment and connectivity diagnostics. This mismatch can mislead operators into approving or invoking a skill with more powerful side effects than its description suggests, increasing the chance of unintended command execution or filesystem changes.

Context-Inappropriate Capability

Medium

Confidence: 92% confidence
Finding: Write capability is not justified by the stated function of a free database guidance skill. Unnecessary write access violates least privilege and could allow the agent to alter local files, configs, or generated scripts when handling ordinary advisory prompts.

Description-Behavior Mismatch

Medium

Confidence: 90% confidence
Finding: The manifest presents the skill as a knowledge-base assistant, but later guidance expands behavior into command execution and host-level connectivity troubleshooting. This creates a capability disclosure gap that may cause users or platform reviewers to underestimate the operational risk of enabling the skill.

Vague Triggers

Medium

Confidence: 82% confidence
Finding: The invocation description is extremely broad, covering general database operations, SQL, and storage management without tight scope boundaries. Overbroad activation increases the chance the skill is selected for many unrelated or higher-risk tasks, including ones where its exec/write permissions become unnecessarily available.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The keyword coverage list is expansive and ambiguous, effectively creating a large activation surface for common database-related phrases. In combination with elevated tools, this can cause the skill to trigger in contexts where users did not intend to permit filesystem or command execution behavior.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The skill exposes exec/write for database diagnostics but does not provide clear upfront warnings about system-impacting actions, despite mentioning ping tests and connectivity checks in troubleshooting guidance. Users may expect only advice from a knowledge-base skill and could unknowingly enable actions affecting the host environment or sensitive network paths.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.