codeql-skill

Security checks across malware telemetry and agentic risk

Overview

This CodeQL skill is mostly coherent, but its scan mode can run local project build commands and is triggered by broad phrases without strong safety scoping.

Install only if you want a CodeQL automation helper and are comfortable with it running local repository scans and project build commands. Use SCAN mode on trusted repositories or in a sandbox/container, and confirm the intended repo path and language before running it.

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: 80% confidence
Finding: The trigger list includes broad phrases such as "scan this repo," "analyze this vulnerability," and "optimize this query," which can cause the skill to activate in contexts where the user did not intend to invoke this specific CodeQL workflow. In a security-focused skill that can drive repository scanning, SARIF handling, and query tuning, accidental invocation can lead to unintended file access, confusing automation behavior, or running the wrong analysis pipeline on sensitive inputs.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal