ClawHub
Back to skill

Security audit

code-review

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed multi-agent code review workflow with proportionate repo and GitHub actions for its stated purpose.

Install this if you want an opinionated, heavier PR review workflow. Be aware that saying ordinary phrases like 'review this PR' may invoke it, and it may add Copilot as a reviewer; confirm before allowing pushes or issue creation, especially in shared repositories.

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
95% confidence
Finding
The file says the trigger is 'review this PR' phrasing, which is a generic request users may say in many ordinary code-review contexts. Because the skill is described as deliberately model-invocable and does not provide exclusion conditions or tighter scope, this creates ambiguity about when this specific multi-agent skill should activate versus a simpler review flow.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.