门店Benchmark分析

Security checks across malware telemetry and agentic risk

Overview

This retail benchmarking skill is purpose-aligned, but it should be reviewed because it queries sensitive store-performance data through an undeclared local database helper and builds SQL from user-controlled inputs.

Review before installing. Use only with read-only, store-scoped database credentials, replace the hardcoded local api_client dependency with a packaged or declared dependency, and require parameterized queries or strict validation for store IDs, scope codes, and dates.

SkillSpector

By NVIDIA

Vulnerability Patterns

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

Findings (1)

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The trigger conditions are very broad and map to common retail-analysis phrases like ranking, comparison, and benchmark requests, which can cause the skill to activate outside its intended scope. Over-broad invocation increases the chance of unintended tool use, misleading answers, or the skill handling requests without sufficient context or user confirmation.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal