What is Opus Deep Analysis?

It is an advanced security check that utilizes a three-agent adversarial pipeline (Red Team, Blue Team, and Auditor) to find and fix complex attack vectors.

How does it handle found vulnerabilities?

The skill provides a detailed report with severity grades (A-F) and can automatically apply safe fixes for issues like hardcoded secrets or wildcard permissions.

What does the Security Scan skill check?

It audits files within the .claude/ directory, specifically checking CLAUDE.md, settings.json, MCP server configurations, and custom hooks for security flaws and secrets.

Can I use this in my CI/CD pipeline?

Yes, it supports JSON and Markdown output formats and can be integrated into GitHub Actions to fail builds if vulnerabilities are detected.

Do I need to install anything extra?

Yes, this skill requires AgentShield (ecc-agentshield), which can be installed via npm or run directly through npx.

Security Scan for Claude Code

Name: Security Scan for Claude Code
Author: oabdelmaksoud

byoabdelmaksoud

0•

보안 및 테스팅

Scans Claude Code project configurations for security vulnerabilities, hardcoded secrets, and prompt injection risks using AgentShield.

This skill integrates AgentShield to perform comprehensive security audits on Claude Code environments, specifically targeting configuration files like CLAUDE.md and settings.json. It identifies critical issues such as unmasked API keys, dangerous shell command permissions, and potential command injection vulnerabilities in hooks or MCP servers. By providing automated fixes and deep analysis through a multi-agent adversarial pipeline, it ensures that your AI-assisted development environment remains secure and compliant with industry best practices.

주요 기능

01Comprehensive scanning of .claude directory files including settings.json and mcp.json

02Multi-agent adversarial analysis for deep security insights (Opus 4.6)

03Detection of hardcoded secrets, API keys, and overly permissive tool access

040 GitHub stars

05Automated remediation for common configuration vulnerabilities and permission issues

06Identification of prompt injection patterns and command injection risks in hooks

사용 사례

01Validating security configurations before committing changes to version control

02Auditing a new repository before starting development with Claude Code

03Performing periodic security hygiene checks on AI agent permissions and MCP server hooks

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add oabdelmaksoud/superclaw security-scan

For use in Claude.ai and ChatGPT

주요 기능

01Comprehensive scanning of .claude directory files including settings.json and mcp.json

02Multi-agent adversarial analysis for deep security insights (Opus 4.6)

03Detection of hardcoded secrets, API keys, and overly permissive tool access

040 GitHub stars

05Automated remediation for common configuration vulnerabilities and permission issues

06Identification of prompt injection patterns and command injection risks in hooks

사용 사례

01Validating security configurations before committing changes to version control

02Auditing a new repository before starting development with Claude Code

03Performing periodic security hygiene checks on AI agent permissions and MCP server hooks