Security audit

SQL生成器(免费版)

Security checks across malware telemetry and agentic risk

Overview

The skill is a SQL helper, but it also enables live database execution and data-changing SQL without clear safety guardrails.

Install only if you are comfortable with an agent helping beyond SQL drafting. Use it with sanitized schema input where possible, avoid production credentials, review every generated statement, and require explicit confirmation before executing INSERT, UPDATE, DELETE, or DDL.

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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

Findings (5)

Description-Behavior Mismatch

Medium

Confidence: 90% confidence
Finding: The skill is presented as a SQL generator, but its documentation broadens behavior into executing generated SQL, using runtime tools, accessing database credentials, and performing host/network diagnostics. That scope expansion increases the chance an agent will move from low-risk content generation into direct system and database actions that can alter data or expose infrastructure details without a tightly justified need.

Context-Inappropriate Capability

Low

Confidence: 95% confidence
Finding: Including instructions to run ping and troubleshoot network connectivity is not justified by the stated purpose of a SQL generation skill. Even if framed as error handling, it encourages unnecessary host-level actions and expands the agent's operational footprint beyond SQL assistance.

Context-Inappropriate Capability

Medium

Confidence: 88% confidence
Finding: The documentation states that database connection credentials are injected via environment variables, which implies the skill may access live database secrets despite being described as a free SQL generator. This is broader than necessary for generating SQL text and creates risk of unintended secret exposure or execution against production-like systems.

Vague Triggers

Medium

Confidence: 82% confidence
Finding: The activation guidance says to use the skill whenever database operations, SQL queries, or data storage management are needed, which is much broader than the narrower description of a SQL generator. Overbroad trigger conditions can cause agents to invoke this skill in inappropriate contexts, increasing the likelihood of unsafe SQL generation or unintended operational actions.

Missing User Warnings

High

Confidence: 97% confidence
Finding: The skill explicitly supports generating INSERT, UPDATE, and DELETE statements but does not prominently warn that these statements can modify or destroy data if executed. In an agent context, omission of that warning is dangerous because users may treat generated SQL as inherently safe and run destructive statements without review or approval.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.