security

Load security patterns from a personal wiki and audit code for vulnerabilities.

Updated Sep 26, 2025

Search Tags:#code review #owasp #threat modeling #security audit #prompt injection #agent sandboxing #wiki knowledge base

Ready to run

npx skills add https://github.com/vietbui1999ru/dotfiles --skill security-vietbui1999ru

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: security
Source: https://github.com/vietbui1999ru/dotfiles/tree/main/llm-wiki-plugin/skills/security
Command: npx skills add https://github.com/vietbui1999ru/dotfiles --skill security-vietbui1999ru

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you systematically load security guidance from your personal wiki so you can review code and designs for vulnerabilities, including AI-specific and agent-specific risks.

Core Features & Use Cases

OWASP-level security auditing: Apply a checklist workflow to flag issues with severity, location, and remediation steps.
AI and agent risk review: Validate agent sandboxing controls and assess exposure to indirect prompt injection.
Security concept grounding: Pull relevant wiki pages for code review, AI pitfalls, and web fingerprinting/proxy concerns.

Quick Start

Load the security patterns from your wiki and review your system design by asking: “Use the security skill to audit my code for OWASP issues and indirect prompt injection, citing the wiki pages you used.”

Frequently Asked Questions about security

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I perform an OWASP security audit on my code using a knowledge base?▼

An OWASP security audit applies a checklist workflow to flag vulnerabilities with severity, location, and remediation steps. This skill queries your wiki knowledge base using lexical and vector search to synthesize a structured vulnerability report with citations.

Can I check my AI agent for indirect prompt injection vulnerabilities?▼

Yes, you can check AI agent designs for indirect prompt injection exposure. The audit process loads security patterns from your personal wiki to validate agent sandboxing controls and identify potential AI-specific input manipulation risks.

What is threat modeling for agent sandboxing and how does it work?▼

Threat modeling for agent sandboxing validates isolation controls against potential escape vectors. It works by querying a wiki knowledge base for relevant security patterns, then synthesizing a checklist-based audit to assess sandbox boundary integrity and remediation steps.

How do I review system designs for security issues with wiki-backed checks?▼

System design security reviews load relevant wiki pages to ground the analysis in established security concepts. The review applies a checklist workflow that flags identified issues with severity ratings, precise locations, remediation guidance, and citations from loaded wiki pages.

Does this security audit require a pre-configured wiki knowledge base?▼

Yes, this security audit requires a pre-configured wiki knowledge base containing security patterns. It functions by querying your personal wiki collection with both lexical and vector search to retrieve the specific security guidance needed for vulnerability-focused reviews.

What limitations exist when auditing code for vulnerabilities without standard dependencies?▼

Auditing code without standard dependencies relies entirely on the quality of your wiki knowledge base content. The audit synthesizes checklist-based vulnerability findings with severity and remediation steps strictly based on the security patterns loaded from your personal wiki pages.