ClawHub

sql-doc

Security checks across malware telemetry and agentic risk

Overview

This skill has a legitimate database-reporting purpose, but it is under-scoped for sensitive police-style data and persists raw incident details to a fixed Word document.

Review before installing. Use only with an authorized, preferably read-only database endpoint; replace the hard-coded sample incident data with user-approved query input; avoid SELECT * on sensitive tables; redact personal identifiers before export; and choose an explicit output location and retention policy for generated reports.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill instructs the agent to send database queries over HTTP to an internal API and then execute a local Python script to process results, but it does not clearly warn about data transmission, sensitivity of returned records, or the side effects of generating documents. In an agent environment, this can lead to unreviewed exfiltration of potentially sensitive data and unsafe execution of downstream processing steps.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The script writes a Word report containing highly sensitive police incident data, including apparent personal identifiers, domestic dispute details, locations, timestamps, and contact information, to a fixed filesystem path without consent, minimization, redaction, or warning. In this skill context, the data is especially sensitive because it concerns law-enforcement incidents and potential personal ID numbers, so creating a persistent document materially increases privacy, compliance, and unauthorized disclosure risk.

VirusTotal

59/59 vendors flagged this skill as clean.

View on VirusTotal