Add 53 Java CWE Security Skills for AI agents

Deterministic remediation skills for Java/Spring Boot security vulnerabilities.
Compatible with skills.sh, Claude Code, Cursor, and other AI assistants.

Categories: Injection, Auth, Crypto, Data Protection, File Handling, Concurrency

Author: fynchauhanuday

README.md

@@ -1,190 +1,57 @@
-# Java CWE Security Skills 🔐
+# Java CWE Security Skills
 
-**java-cwe-security-skills** is an open-source library of **deterministic security remediation skills for Java applications**, mapped directly to the **MITRE Common Weakness Enumeration (CWE)** framework.
+[![skills.sh](https://img.shields.io/badge/skills.sh-indexed-blue)](https://skills.sh)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 
-The repository provides structured **`SKILL.md` definitions** that enable AI coding assistants and security agents to **detect, analyze, and fix vulnerabilities automatically** in Java codebases.
+> **53 deterministic AI-agent remediation skills for Java security vulnerabilities**
 
-These skills are designed for integration with modern **AI developer platforms**, including:
+## 🎯 Purpose
 
-* **Augment Code**
-* **skills.sh**
-* **Claude Code**
-* **Cursor AI**
-* **LobeHub Skills Marketplace**
-* **DevSecOps automation pipelines**
+This repository provides **structured remediation skills** that teach AI coding assistants exactly how to fix security vulnerabilities with real code transformations.
 
----
-
-# Why This Project Exists
-
-Modern development teams rely heavily on **SAST tools** such as:
-
-* Checkmarx
-* SonarQube
-* Snyk
-* CodeQL
-* OWASP Dependency Check
+## 🔧 Compatible With
 
-While these tools identify vulnerabilities, they often lack **deterministic remediation patterns**.
+- [skills.sh](https://skills.sh) marketplace
+- Claude Code
+- Cursor
+- LobeHub
+- Other AI coding assistants
 
-This repository bridges that gap by providing **AI-readable remediation skills** that allow coding agents to:
+## 📋 Skills
 
-1. Detect vulnerable code patterns
-2. Map them to **specific CWE weaknesses**
-3. Apply **secure code transformations**
-4. Verify the fix through deterministic rules
+See [index.md](index.md) for the complete list of skills.
 
-This enables **AI-assisted vulnerability remediation** directly inside development workflows.
+## 🚀 Quick Start
 
----
-
-# Features
+Reference a skill when asking AI to fix vulnerabilities:
 
-✔ Deterministic remediation for **Java security vulnerabilities**
-✔ Coverage for **100+ Java-relevant CWE weaknesses**
-✔ Designed for **AI coding assistants and security agents**
-✔ Compatible with **Augment Code skill workflows**
-✔ Supports **automated vulnerability fixing in repositories**
-✔ Based on **MITRE CWE, OWASP, CodeQL, and secure coding best practices**
-
----
+```
+Fix the SQL Injection vulnerability using the cwe-89-sql-injection skill
+```
 
-# Repository Structure
+## 📦 Repository Structure
 
-```id="7x4ihr"
-java-cwe-security-skills
-│
-├── cwe-79-xss
-│   └── SKILL.md
-│
-├── cwe-89-sql-injection
-│   └── SKILL.md
-│
-├── cwe-611-xxe
+```
+├── cwe-89-sql-injection/
 │   └── SKILL.md
-│
-├── cwe-918-ssrf
+├── cwe-79-xss/
 │   └── SKILL.md
-│
-├── cwe-502-insecure-deserialization
+├── cwe-611-xxe/
 │   └── SKILL.md
-│
-└── ...
+└── index.md
 ```
 
-Each folder represents a **single vulnerability remediation skill**.
-
----
-
-# Example Skill
-
-Example: **CWE-89 SQL Injection**
+## 🤝 Contributing
 
-The skill defines:
+Pull requests welcome! Add new skills following the existing format.
 
-* vulnerable Java patterns
-* deterministic remediation logic
-* secure implementation examples
-* verification rules
+## 📚 Sources
 
-AI coding assistants can automatically transform vulnerable SQL queries into **secure prepared statements**.
+- MITRE CWE Database
+- OWASP Cheat Sheets
+- GitHub CodeQL queries
+- VulnerableApp dataset
 
 ---
 
-# Supported Vulnerability Categories
-
-The library includes remediation skills for major vulnerability classes.
-
-### Injection
-
-* SQL Injection (CWE-89)
-* LDAP Injection (CWE-90)
-* XPath Injection (CWE-643)
-* Expression Language Injection (CWE-917)
-
-### Web Security
-
-* Cross-Site Scripting (CWE-79)
-* HTTP Response Splitting (CWE-113)
-* Clickjacking / Missing Security Headers (CWE-693)
-
-### Cryptography
-
-* Weak Cryptography (CWE-327)
-* Hardcoded Cryptographic Keys (CWE-321)
-* Missing Random IV (CWE-329)
-
-### Data Exposure
-
-* Information Exposure (CWE-200)
-* Sensitive Data in Logs (CWE-532)
-* Privacy Violations (CWE-359)
-
-### Deserialization
-
-* Insecure Deserialization (CWE-502)
-
-### Resource Management
-
-* Resource Exhaustion (CWE-400)
-* Uncontrolled Memory Allocation (CWE-789)
-
-### Input Validation
-
-* Path Traversal (CWE-22)
-* Regex Denial of Service (CWE-1333)
-* Unchecked Loop Condition (CWE-606)
-
----
-
-# Using These Skills with Augment Code
-
-Augment Code can use these skills to perform **AI-assisted vulnerability remediation** during development.
-
-Typical workflow:
-
-1. Augment scans the repository for vulnerable patterns
-2. The vulnerability is mapped to a **CWE skill**
-3. The skill provides **deterministic remediation steps**
-4. Augment generates a **secure code transformation**
-
-Example prompt inside Augment:
-
-```
-Fix SQL injection vulnerabilities in this Java service.
-```
-
-The agent loads the **CWE-89 skill** and applies the secure remediation.
-
----
-
-# Sources
-
-Security knowledge in this repository is derived from:
-
-* MITRE CWE Database
-* OWASP Security Cheat Sheets
-* GitHub CodeQL Security Queries
-* Snyk Vulnerability Database
-* OWASP VulnerableApp
-
----
-
-# Contributing
-
-Contributions are welcome.
-
-You can contribute by:
-
-* adding new CWE skills
-* improving remediation patterns
-* adding secure code examples
-* expanding coverage for Java frameworks
-
----
-
-# Vision
-
-The goal of this project is to create the **largest open-source library of AI security remediation skills for Java**.
-
-By combining **CWE knowledge + deterministic remediation patterns**, AI agents can evolve from **vulnerability detection → automatic secure code remediation**.
+**Generated**: 2026-03-07