skills.rest
Vibecoder79
Vibecoder79OFFICIAL
Explore all skills by this author →
WebsiteGitHub

security-architect

Performs threat modeling, code security reviews, audits, and prompt-injection scans for Claude Code skills.

2|Updated Apr 17, 2026
Search Tags:#code review#security#owasp#threat modeling#security audit#stride#prompt injection
One-click install
npx skills add https://github.com/Vibecoder79/jarvis --skill security-architect-vibecoder79
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: security-architect
Source: https://github.com/Vibecoder79/jarvis/tree/main/.claude/skills/security-architect
Command: npx skills add https://github.com/Vibecoder79/jarvis --skill security-architect-vibecoder79

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Security is often skipped or checked only at the end of development. This Skill embeds security engineering into every phase of the Claude Code workflow — threat modeling before code is written, reviews while code changes, full audits before releases, and prompt-injection scanning before installing third-party skills. ## Core Features & Use Cases - DESIGN mode: STRIDE threat modeling with DREAD risk scoring (1-10) during planning, producing prioritized security requirements. - REVIEW mode: OWASP Top 10:2025 checks against code diffs, language-specific secure code patterns (JS/TS, Python, Go, Rust, Java, PHP, C/C++, Bash), secrets detection, and security header audits. - AUDIT mode: Full-project scans covering dependencies, configuration, attack surface mapping, and Agentic AI security (OWASP ASI01-ASI10). - SKILL-SCAN mode: Detects prompt injection attacks in external SKILL.md files across 8 attack categories (override, exfiltration, privilege escalation, destructive actions, settings manipulation, indirect injection, hidden instructions, social engineering) before installation. - Use Case: Before installing a skill from GitHub, run a SKILL-SCAN to verify it does not read your SSH keys, modify CLAUDE.md, or contain hidden instructions. ## Quick Start Ask Claude to run a security review of your recent code changes or scan a skill before installing it.

Frequently Asked Questions about security-architect

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

FAQPage Schema
How do I scan a Claude Code skill for prompt injection before installing it?▼

Use the SKILL-SCAN mode by saying "scan this skill" or "pruefe diesen skill" and providing the SKILL.md file. It checks 8 attack categories including exfiltration, override attempts, and hidden instructions, then returns a SAFE, SUSPICIOUS, or DANGEROUS verdict.

How do I create a threat model with STRIDE and DREAD?▼

Trigger the DESIGN mode during planning by asking for a threat model. The skill defines system scope and trust boundaries, runs STRIDE analysis per component, scores each threat with DREAD (1-10), and outputs prioritized security requirements.

What does the OWASP Top 10:2025 code review check?▼

The REVIEW mode checks code changes against all ten OWASP categories, from broken access control to exception handling. It also verifies language-specific secure patterns, scans for hardcoded secrets, and audits security headers for web applications.

Which programming languages does the security review support?▼

The secure code pattern references cover JavaScript/TypeScript, Python, Go, Rust, Java, PHP, C/C++, and Bash. Each language section lists unsafe patterns, safe alternatives, and a watchlist of dangerous functions.

Does the security audit cover AI agents and MCP servers?▼

Yes, the AUDIT mode includes Agentic AI Security based on OWASP ASI01-ASI10. It covers goal hijacking, tool misuse, memory poisoning, and rogue agents, with a checklist for projects using AI agents or tool-calling systems.

When should I not rely on this skill for security?▼

The skill performs static, instruction-based analysis and does not execute dynamic penetration tests or runtime monitoring. For critical systems, combine its findings with dedicated SAST/DAST tooling and professional penetration testing.

Related Skills

Explore More Software Engineering Skills→
davila7davila7
30.5k

threat-modeling-expert

Analyzes system architectures using STRIDE and attack trees to identify and prioritize security threats.

Community
Intermediate
TravisBumgarnerTravisBumgarner

ralph-refactor

Analyze code and tests to identify quality issues and security gaps.

Community
Advanced
EtroxTaranEtroxTaran

security-scan

Automates security audits combining SAST, MLOps-driven fleet monitoring, and secret detection for GitOps workflows.

Community
Advanced
TECHKNOWMAD-LABSTECHKNOWMAD-LABS

security-audit

Scan code repositories for vulnerabilities and leaked secrets using static analysis tools.

Official
Advanced