security-scan

Scan Claude Code configurations for security vulnerabilities and misconfigurations.

4|Updated Feb 14, 2026
One-click install
npx skills add https://github.com/KuaaMU/omnihive --skill security-scan-kuaamu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: security-scan
Source: https://github.com/KuaaMU/omnihive/tree/main/library/ecc-skills/security-scan
Command: npx skills add https://github.com/KuaaMU/omnihive --skill security-scan-kuaamu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill addresses the critical need to identify and mitigate security vulnerabilities, misconfigurations, and potential injection risks within Claude Code project configurations, ensuring a safer development environment.

Core Features & Use Cases

  • Comprehensive Auditing: Scans CLAUDE.md, settings.json, MCP servers, hooks, and agent definitions for security flaws.
  • Vulnerability Detection: Identifies hardcoded secrets, prompt injection patterns, overly permissive access controls, and risky server configurations.
  • Use Case: Before deploying a new Claude Code project or after making significant configuration changes, run this scan to receive a security grade and actionable recommendations to harden your setup.

Quick Start

Run the security scan against the current project's .claude/ directory.

Frequently Asked Questions about security-scan

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

FAQPage Schema
How do I scan my Claude Code configuration for security vulnerabilities?

Scanning a Claude Code configuration involves running an AgentShield scan against the `.claude/` directory to analyze `CLAUDE.md`, `settings.json`, and MCP servers for hardcoded secrets and injection patterns, resulting in a security grade and actionable recommendations.

What security risks exist in Claude Code MCP servers and hooks?

Security risks in Claude Code MCP servers and hooks include hardcoded secrets, prompt injection patterns, overly permissive access controls, and risky server configurations that expose your development environment to unauthorized access.

When should I run a security audit on my Claude Code project?

Run a security audit on a Claude Code project before deploying a new project or after making significant configuration changes to ensure ongoing security hygiene and identify newly introduced misconfigurations.

Does AgentShield work with existing settings.json and agent definitions?

AgentShield works with existing `settings.json` and agent definitions by scanning them for overly permissive access and risky configurations, evaluating your current setup without requiring modifications before the initial scan.

What is the best way to secure Claude Code configurations against prompt injection?

The best way to secure Claude Code configurations against prompt injection is to run a comprehensive scan that analyzes agent definitions and `CLAUDE.md` files for injection patterns, providing actionable recommendations to harden your setup.