来源信息
- 仓库
- ApiliumCode/mayros
- 最近来源活动
- 2026年2月26日 12:53
- 检测到的 SKILL.md 语言
- 英语
- 星标
- 13
- 分支
- 1
安装方式
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
检查来源文件
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
文件资源管理器
10 个文件正在显示 SKILL.md
SKILL.md
来源说明 · 只读预览- name
- code-security
- description
- Classifies code vulnerabilities by OWASP Top 10 and CWE identifiers
- type
- semantic
- user-invocable
- true
- semantic
- {"skillVersion":1,"permissions":{"graph":["read","write"],"proofs":["request"],"memory":["recall"]},"assertions":[{"predicate":"security:vulnerability","requireProof":true},{"predicate":"security:finding","requireProof":false}],"queries":[{"predicate":"security:history","scope":"agent"}]}
# code-security
Classifies code vulnerabilities by OWASP Top 10 categories and CWE identifiers, providing severity ratings for each finding.
## When to Use
Use this skill when:
- Reviewing code snippets stored in the knowledge graph for security vulnerabilities
- Performing automated security audits on code content
- Classifying findings by CWE identifier and severity level before human review
## Vulnerability Coverage
The skill detects 12 vulnerability classes:
| CWE | Category | Severity |
| ------- | ---------------------------------- | -------- |
| CWE-89 | SQL Injection | critical |
| CWE-79 | Cross-Site Scripting (XSS) | high |
| CWE-78 | Command Injection | critical |
| CWE-22 | Path Traversal | high |
| CWE-502 | Insecure Deserialization | high |
| CWE-798 | Hardcoded Secrets | critical |
| CWE-95 | Code Injection | critical |
| CWE-918 | Server-Side Request Forgery (SSRF) | high |
| CWE-327 | Weak Cryptography | medium |
| CWE-601 | Open Redirect | medium |
| CWE-209 | Information Exposure | low |
| CWE-306 | Missing Authentication | medium |
## Instructions
1. Store code content in the graph using `skill_assert` or other write tools
2. Query with predicate `security:finding` to trigger the code-security runtime
3. The skill enriches each result with a `findings` array containing CWE IDs and severity levels
4. Use `security:vulnerability` (with proof) to record confirmed vulnerabilities
5. Query `security:history` to review past audit results for the current agent
在 GitHub 查看