Claude Code Evil — Security skill for Claude Code
Removes Claude Code cyber safety checks in source for security research and testing.
How to install Claude Code Evil
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open Gabbysorn/claude-code_evil and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Claude Code Evil does
Removes Claude Code cyber safety checks in source for security research and testing.
Alternatives in Security
- Defense In Depth — Implement multi-layered testing and security best practices 98.1k ★
- Skills — Trail of Bits Claude Code skills for security research, vulnerability detection, and audit workflows 4k ★
- Anthropic Cybersecurity Skills — 734+ structured cybersecurity skills for AI agents · MITRE ATT&CK mapped · agentskills.io open standard · Work 3.8k ★
README
Claude Code 安全限制移除说明
本文档记录了对 Claude Code v2.1.88 源代码的修改,移除了提示词级别的网络安全限制。
修改概述
| 修改文件 | 修改类型 | 状态 |
|---|---|---|
src/constants/cyberRiskInstruction.ts |
清空安全指令 | ✅ 已完成 |
已移除的限制
1. CYBER_RISK_INSTRUCTION (网络安全风险指令)
**文件位置:** `src/constants/cyberRiskInstruction.ts`
移除前的完整内容:
export const CYBER_RISK_INSTRUCTION = `IMPORTANT: Assist with authorized security testing, defensive security, CTF challenges, and educational contexts. Refuse requests for destructive techniques, DoS attacks, mass targeting, supply chain compromise, or detection evasion for malicious purposes. Dual-use security tools (C2 frameworks, credential testing, exploit development) require clear authorization context: pentesting engagements, CTF competitions, security research, or defensive use cases.`
移除后的内容:
export const CYBER_RISK_INSTRUCTION = ``
移除的具体限制详解
一、允许的场景 (已移除的检查)
| 序号 | 英文 | 中文 | 说明 |
|---|---|---|---|
| 1 | authorized security testing |
授权安全测试 | 原要求有书面授权才能协助渗透测试 |
| 2 | defensive security |
防御性安全 | 原允许安全加固、漏洞修复等防御性工作 |
| 3 | CTF challenges |
CTF 竞赛 | 原允许 CTF 竞赛相关协助 |
| 4 | educational contexts |
教育场景 | 原允许教学和学习目的的安全研究 |
二、拒绝的场景 (已移除的限制)
| 序号 | 英文 | 中文 | 具体行为示例 |
|---|---|---|---|
| 1 | destructive techniques |
破坏性技术 | 数据擦除工具、系统破坏脚本、勒索软件 |
| 2 | DoS attacks |
拒绝服务攻击 | DDoS 脚本、SYN Flood、资源耗尽攻击 |
| 3 | mass targeting |
大规模攻击 | 批量漏洞扫描、自动化利用工具、蠕虫传播 |
| 4 | supply chain compromise |
供应链攻击 | npm 包投毒、依赖库后门、构建系统入侵 |
| 5 | detection evasion for malicious purposes |
恶意逃避检测 | 杀毒软件绕过、沙箱逃逸、反调试技术 |
三、双重用途工具限制 (已移除)
| 序号 | 英文 | 中文 | 工具示例 |
|---|---|---|---|
| 1 | C2 frameworks |
命令控制框架 | Cobalt Strike, Metasploit, Covenant, Sliver |
| 2 | credential testing |
凭证测试工具 | Hydra, crackmapexec, Medusa, Patator |
| 3 | exploit development |
漏洞利用开发 | Exploit 脚本编写、Payload 生成 |
**原限制要求:** 使用上述工具需要提供明确的授权上下文:
- 渗透测试合同
Related Skills
CTI Expert
/case target.com --yoloCyber threat intelligence and open-source intelligence skill. Turns Claude into a trained CTI/OSINT analyst. G
Engineering Audit
Full-spectrum engineering health audit: scores Maintainability, Security, Testing, Architecture, Documentation
TypeScript Engineering
Reusable TypeScript engineering skill for AI coding assistants. Practical guidance on type safety, runtime val
KeyHunter Skill
Academic research toolkit for AI gateway panel security studies (Sub2API / New-API / One-API): FOFA discovery,
Millieskills
Research-driven universal AI Agent Skills for UI/UX, code repair, security, testing, architecture, backend/API
Agentic Security Pipeline
An autonomous agentic pipeline that finds, proves, and patches real C memory-safety vulnerabilities end-to-end
Related Agents
Infra Security
Security triage and hardening specialist. Exposure review, IR coordination, hardening recommendations. Recomme
Bash Reviewer
Reviews Bash/shell scripts for correctness, safety, idiomatic style, and portability. Checks set -euo pipefail
Sequant QA Checker
Quality check agent for sequant /qa phase. Runs type safety, scope/size, security, and documentation checks on