skills.rest
jjmendezrodriguez
jjmendezrodriguezCOMMUNITY
Explore all skills by this author →
GitHub

security-scan

Scan Claude Code configuration files for security vulnerabilities and misconfigurations.

Updated Apr 16, 2026
Search Tags:#claude code#hooks#security audit#mcp#prompt injection#secrets scanning#ci hardening
One-click install
npx skills add https://github.com/jjmendezrodriguez/jm-claude-plugin --skill security-scan-jjmendezrodriguez
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: security-scan
Source: https://github.com/jjmendezrodriguez/jm-claude-plugin/tree/main/skills/security-scan
Command: npx skills add https://github.com/jjmendezrodriguez/jm-claude-plugin --skill security-scan-jjmendezrodriguez

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires npx ecc-agentshield, AgentShield.

What problem does it solve?

It helps prevent security breaches caused by unsafe or misconfigured Claude Code settings, prompts, MCP server definitions, and hooks by identifying common injection and secrets risks before they reach production.

Core Features & Use Cases

  • Configuration Security Auditing: Scans CLAUDE.md, .claude/settings.json, mcp.json, and agent definitions for hardcoded secrets, overly permissive permissions, and prompt-injection patterns.
  • Hook & MCP Risk Detection: Flags command-injection vectors in hooks (including interpolation risks) and identifies risky MCP servers and supply-chain issues.
  • Reportable, CI-Friendly Outputs: Produces terminal, JSON, Markdown, or HTML reports suitable for local review or automated pipelines.
  • Optional Auto-Fix for Safe Changes: Applies only auto-fixable remediations such as replacing hardcoded secrets with environment references and tightening wildcard permissions.

Quick Start

Run a medium-severity scan over the current project’s .claude directory using AgentShield.

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?▼

You can scan Claude Code configuration files for security vulnerabilities by running AgentShield to detect misconfigurations, hardcoded secrets, and prompt injection risks in CLAUDE.md and .claude/settings.json.

What is prompt injection detection in MCP server definitions?▼

Prompt injection detection in MCP server definitions identifies risky command-injection vectors and supply-chain issues within your MCP config and hooks before they reach production environments.

How do I audit Claude Code hooks for command injection risks?▼

You audit Claude Code hooks for command injection risks by scanning hook configurations to flag interpolation vulnerabilities and unsafe command execution patterns using AgentShield.

Can I auto-fix hardcoded secrets and wildcard permissions in .claude settings?▼

Yes, you can auto-fix hardcoded secrets and wildcard permissions in .claude settings by applying safe remediations that replace secrets with environment references and tighten overly permissive access.

Does the security scan support JSON or HTML reports for CI pipelines?▼

Yes, the security scan supports JSON, Markdown, and HTML report outputs suitable for local review or integration into automated CI pipelines for continuous configuration hardening.

Do I need AgentShield to run a security audit on CLAUDE.md files?▼

Yes, you need AgentShield to run a security audit on CLAUDE.md files, as the Skill uses the npx ecc-agentshield package to produce actionable findings and apply safe auto-fixes.

Related Skills

Explore More Software Engineering Skills→
evolv3-aievolv3-ai
1

admin-mcp

Install and configure MCP servers for Claude Desktop on Windows and WSL contexts.

Official
Advanced
sgcarstrendssgcarstrends
21

package-audit

Identify security vulnerabilities in npm dependencies using pnpm audit and Snyk.

Official
Advanced
ValenBorrasValenBorras
1

security-scan

Scan Claude Code configurations for security vulnerabilities and misconfigurations.

Community
Advanced
brainbrewlabsbrainbrewlabs
7

code-fixing

Fix code scanner, security auditor, and linter issues with verified edits and reports.

Official
Intermediate