ClawHub
Back to skill

Security audit

NEXUS Sql Builder

Security checks across malware telemetry and agentic risk

Overview

This SQL helper is not clearly malicious, but it can automatically send prompts and payment credentials to a third-party paid service without well-defined consent boundaries.

Install only if you trust NEXUS with SQL requirements, schema details, prompts, and payment proofs. Use sandbox or narrowly scoped payment credentials where possible, avoid secrets or regulated data, and configure your agent to require explicit approval before paid remote calls.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The README states the skill is 'automatically invoked' when a matching task is detected, but it does not define what constitutes a match, what data may be sent, or whether user confirmation is required. In an agent ecosystem, vague auto-invocation boundaries can cause unintended activation and remote handling of sensitive prompts, increasing the risk of unauthorized data exposure or unexpected paid actions.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The README instructs users or agents to POST arbitrary 'input' to a third-party remote endpoint but provides no warning that prompts, schema details, business logic, or sensitive data may leave the local environment. For a SQL-building skill, submitted requirements may easily contain proprietary database structure or confidential query intent, so omission of a privacy/data-transmission warning is a meaningful security issue.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.