security-scan

Scan Claude Code configuration files for security vulnerabilities and misconfigurations.

Updated May 9, 2026
One-click install
npx skills add https://github.com/kk20300113-png/my-claude-skills --skill security-scan-kk20300113-png
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: security-scan
Source: https://github.com/kk20300113-png/my-claude-skills/tree/main/security-scan
Command: npx skills add https://github.com/kk20300113-png/my-claude-skills --skill security-scan-kk20300113-png

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Manually reviewing Claude Code configuration files for security flaws is time-consuming and error-prone, with risks like hardcoded secrets, overly permissive access rules, and prompt injection vectors often going unnoticed until they cause a breach.

Core Features & Use Cases

  • Comprehensive Config Scanning: Checks all core Claude Code files including CLAUDE.md, settings.json, MCP server configs, hooks, and agent definitions for vulnerabilities.
  • Severity-Graded Reporting: Categorizes findings by criticality with clear remediation guidance, and offers safe auto-fixes for common low-risk issues.
  • Use Case: Before committing changes to your project's Claude Code setup, run this scan to catch hardcoded API keys and unrestricted shell access permissions that could expose sensitive data.

Quick Start

Use the security-scan skill to run a full security audit of your project's Claude Code configuration and identify any critical vulnerabilities.

Frequently Asked Questions about security-scan

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

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

Scanning Claude Code configuration files for security vulnerabilities involves validating CLAUDE.md, settings.json, and MCP configs against best practices using AgentShield. This detects hardcoded secrets, overly permissive access rules, and prompt injection vectors before they cause breaches.

What security risks should I check for in CLAUDE.md and settings.json before committing?

Checking CLAUDE.md and settings.json for security risks involves looking for hardcoded API keys, unrestricted shell access permissions, and prompt injection vectors. Scanning these configuration files detects misconfigurations and sensitive data exposure risks before commit.

Can I use a security scan to check Claude Code MCP server configs and hooks?

Yes, a configuration security scan can check Claude Code MCP server configs and hooks. The scan applies AgentShield validation to these files to detect misconfigurations, injection risks, and overly permissive access rules across your project setup.

How do I integrate a Claude Code security audit into a CI/CD pipeline gate?

Integrating a Claude Code security audit into a CI/CD pipeline gate involves running a configuration scan during pipeline execution. The scan validates setup files and generates severity-graded reports with remediation guidance, acting as a security gate to block unsafe deployments.

Does the Claude Code vulnerability scan provide remediation guidance for detected issues?

Yes, the Claude Code vulnerability scan provides remediation guidance for detected issues. It delivers severity-graded reporting that categorizes findings by criticality and offers safe auto-fixes for common low-risk misconfigurations found in your settings.