ClawHub
Back to skill

Security audit

Db Explorer

Security checks across malware telemetry and agentic risk

Overview

This database helper is sensitive because it can use credentials and run database commands, but its behavior is disclosed and aligned with its stated purpose.

Install only if you want an agent to help access databases from your environment. Use read-only or least-privilege non-production credentials when possible, review every command before it runs, explicitly approve any restore/import/migration/write, avoid broad production scans, and delete exported or backup files that may contain sensitive data.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The skill's documented capabilities expand from exploration/querying into backup, restore, and migration workflows that can alter or overwrite data and move data across systems. In an agent setting, this scope creep is dangerous because users may invoke a seemingly read-oriented skill and the agent may select destructive or high-impact commands that exceed the declared purpose.

Intent-Code Divergence

Medium
Confidence
96% confidence
Finding
The document states the skill is read-only by default, but later includes restore and data-loading commands that perform writes without the same confirmation and transaction framing. This inconsistency can mislead downstream agents or users into trusting the skill as safe while it contains operational guidance for destructive changes.

Vague Triggers

Medium
Confidence
85% confidence
Finding
The activation triggers are broad enough to match common requests like 'check the database' or 'show me the data,' which can cause the skill to activate in loosely related contexts. In an automated agent environment, overbroad routing increases the chance that powerful database tooling is selected unnecessarily, exposing credentials, schemas, or live systems when a safer or narrower skill would suffice.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The export, backup, restore, and migration instructions create files and may move or overwrite sensitive data, but they lack prominent warnings about data exposure, local file permissions, destination safety, and modification risk. In practice this can lead to accidental leakage of production data to world-readable temp paths or unintended writes to databases and local storage.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.