ClawHub
Back to skill

Security audit

Sql Query Tool Free

Security checks across malware telemetry and agentic risk

Overview

The skill is a coherent SQL helper, but it enables broad command-line database execution and write/delete-style operations without clear safeguards for real databases.

Install only if you are comfortable with an agent using command-line database tools. Treat it as capable of changing real database state; prefer read-only credentials, test databases, explicit per-command approval, backups, and transactions with rollback plans before allowing writes or deletes.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Vague Triggers

Medium
Confidence
82% confidence
Finding
The activation guidance is broad enough to trigger on many generic database or storage tasks, while the skill exposes read/write/exec capabilities. Over-broad invocation can cause an agent to select this skill in situations involving production data or state-changing operations without sufficient scoping, increasing the chance of unsafe execution.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill advertises create/query/modify/delete behavior and lists write plus exec tools, but it does not prominently warn about persistent data modification risk. In an agent context, this can normalize state-changing actions and lead to unintended writes, deletions, or operational impact against real databases.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The example shows state-changing SQL within a transaction without an explicit warning that it alters persistent data and may be irreversible if run against real systems. In a skill intended for AI agents with exec capability, such examples can be copied into production-like environments and cause unintended account balance changes or similar data integrity issues.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.