Security audit

NEXUS Text To Sql

Security checks across malware telemetry and agentic risk

Overview

This is a coherent paid remote text-to-SQL skill, but it should be reviewed because automatic use could send prompts to a third-party service and spend payment credentials without clear per-call controls.

Install only if you trust NEXUS and are comfortable sending natural-language prompts, schema details, and payment proof credentials to its hosted service. Use sandbox_test first, avoid secrets or private database data in prompts, and configure agent-level confirmation or spending limits before enabling automatic invocation.

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: 89% confidence
Finding: The README states the skill is 'automatically invoked' when a matching task is detected, but provides no trigger boundaries, consent model, or safeguards. In a skill that sends prompts to a remote paid endpoint, ambiguous auto-invocation increases the risk of unintended data disclosure and unexpected billing.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The README demonstrates sending arbitrary input to a third-party hosted API and describes pricing, but does not clearly warn that user prompts may leave the local environment and be processed by a remote paid service. This omission can cause users or downstream agents to transmit sensitive data without informed consent and incur unanticipated cost.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.