by zhaji2333
基于 Claude Code / Codex 的 SRC 漏洞挖掘 Agent 技能体系 —— 系统级提示词 + 14 个专项安全测试 Skill 知识库
# Add to your Claude Code skills
git clone https://github.com/zhaji2333/CkSKILLSCkSKILLS is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by zhaji2333. 基于 Claude Code / Codex 的 SRC 漏洞挖掘 Agent 技能体系 —— 系统级提示词 + 14 个专项安全测试 Skill 知识库. It has 50 GitHub stars.
CkSKILLS's catalog security scan is still queued. You can run an instant dependency and prompt-injection check now with the "Scan for vulnerabilities" button above.
Clone the repository with "git clone https://github.com/zhaji2333/CkSKILLS" and add it to your Claude Code skills directory (see the Installation section above).
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh CkSKILLS against similar tools.
No comments yet. Be the first to share your thoughts!
Unlocks once the catalog security scan passes (runs nightly).
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
The deep catalog scan for this skill is still queued. Run an instant dependency check now instead.
基于 Claude Code / Codex 的 SRC 漏洞挖掘 Agent 技能体系 —— 将顶尖安全研究员的方法论沉淀为可调度、可复用的 Skill 知识资产。
CK-Skills 是一套面向 SRC 漏洞挖掘的 Agent 提示词工程与技能知识体系,以 Claude Code / Codex 等通用 Agent 框架为运行时,通过 系统级提示词(AGENTS.md)+ 14 个专项技能知识库(Skills),让通用 Agent 在垂直安全领域达到专家级表现。
针对通用 Agent 在垂直领域的三大痛点:
四层架构,各司其职:
┌──────────────────────────────────────────────┐
│ Agent 运行时(Claude Code / Codex) │ 推理与工具调用
├──────────────────────────────────────────────┤
│ 系统级提示词 AGENTS.md │ 身份 / 约束 / 纪律 / 调度
├──────────────────────────────────────────────┤
│ 14 个专项 Skills(知识层) │ 方法论 / 场景表 / 步骤
├──────────────────────────────────────────────┤
│ 触发路由(场景→技能 / 漏洞类型→技能) │ 专家知识按需加载
└──────────────────────────────────────────────┘
| 技能 | 覆盖范围 |
|---|---|
recon-js-analysis |
资产测绘、webpack / source map 还原、API 与密钥提取 |
auth-access-control |
认证绕过、越权、IDOR、多租户隔离、密码重置、JWT |
injection-vulns |
SQL / NoSQL / 命令 / SSTI / 表达式注入 |
business-logic-race |
支付逻辑、状态机建模、金额篡改、竞态条件 |
file-handling |
文件上传 getshell、路径穿越、Zip Slip、CSV 注入 |
ssrf-internal-network |
SSRF、云元数据、内网探测、DNS 重绑定 |
deserialization-xxe |
反序列化 RCE、XXE、原型污染、利用链构造 |
xss-frontend-security |
XSS、CSRF、CORS、Clickjacking |
api-protocol-security |
BOLA、GraphQL、WebSocket、HTTP 走私 |
mobile-iot-device-security |
Android / iOS 逆向、WebView、固件安全 |
cloud-infra-supply-chain |
云配置错误、K8s、CI/CD、SBOM、供应链 |
source-code-audit |
输入点 → 传播链 → Sink 静态审计 |
waf-bypass-techniques |
Level 1-7 对抗升级框架 |
ai-llm-agent-security |
提示词注入、越狱逃逸、System Prompt 泄露、RAG/记忆污染、Agent 工具滥用致 RCE/SSRF、沙箱逃逸、模型供应链 |
将本项目 .agents/ 目录与 AGENTS.md 放入你的工作目录,Agent 会自动加载 AGENTS.md 作为系统提示词,并根据触发信号加载对应 Skill。
CK-Skills/
├── AGENTS.md # 系统级提示词总纲
├── .agents/
│ └── skills/ # 14 个专项技能
│ ├── recon-js-analysis/
│ │ └── SKILL.md
│ ├── auth-access-control/
│ │ └── SKILL.md
│ └── ... # 其余专项技能
在 Claude Code / Codex 中直接描述测试场景,Agent 会根据触发信号自动加载对应 Skill 并按方法论执行:
目标:https://example.com,已登录普通用户,需要测试越权
→ 自动加载 auth-access-control,按"权限三问"执行
欢迎提交 Issue / PR 补充 Skill 或完善方法论。新增 Skill 只需按现有 DSL 编写 SKILL.md 并在 AGENTS.md 路由表注册。
本项目仅用于授权安全测试、安全研究与教学。使用者需遵守所在国家/地区的法律法规,在获得目标所有者明确书面授权的前提下使用。禁止用于任何非法用途。使用者自行承担因不当使用产生的一切法律责任。