security-check

Assess third-party skills, agents, or plugins for security vulnerabilities.

Updated May 17, 2026
One-click install
npx skills add https://github.com/ievo-ai/skills --skill security-check-ievo-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: security-check
Source: https://github.com/ievo-ai/skills/tree/main/plugins/ievo/skills/security-check
Command: npx skills add https://github.com/ievo-ai/skills --skill security-check-ievo-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires gh, git, WebFetch, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill ensures the safety of third-party skills, agents, or plugins before installation by conducting a thorough vulnerability assessment by a senior application security engineer.

Core Features & Use Cases

  • Vulnerability Assessment: Comprehensive scanning for potential security issues such as prompt injection, credential exfiltration, and supply-chain compromise.
  • Deep Content Review: Inspects the full content of every file, including dependencies, to ensure no malicious code is present.
  • Structured Verdict: Provides a GREEN/YELLOW/RED verdict with cited evidence and alternative suggestions if necessary.

Quick Start

Run the security-check skill with the following command: /ievo:security-check <candidate> <type>

Frequently Asked Questions about security-check

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a vulnerability assessment on third-party plugins before installation?

Run a vulnerability assessment on third-party skills by executing the security-check command with the candidate name and type. It clones the candidate via git, inspects all files and dependencies, and outputs a structured security verdict with cited evidence.

What security issues are detected during a deep content code review?

A deep content code review detects prompt injection, credential exfiltration, and supply-chain compromise threats. It examines the full content of every file, including dependencies, to ensure no malicious code is present before providing a structured safety verdict.

Do I need gh CLI and git installed to perform a security scan on agents?

Yes, you need gh CLI and git installed to perform a security scan on agents. The gh CLI retrieves API metadata, git clones the candidate repository for local file reads, and WebFetch pulls additional audit signals during the vulnerability assessment.

What's the best way to check application security and prevent prompt injection in new skills?

The best way to check application security and prevent prompt injection is using an automated vulnerability assessment. It audits new skills by examining code and metadata, catching injection attempts, and providing remediation suggestions with a clear safety verdict.

What does the GREEN, YELLOW, or RED threat detection verdict mean for code safety?

The threat detection verdict indicates code safety levels: GREEN means safe, YELLOW flags potential issues requiring review, and RED identifies critical security vulnerabilities. The structured report includes cited evidence and alternative suggestions if necessary.

Related Skills