skills.rest
jassics
jassicsCOMMUNITY
Explore all skills by this author →
Website𝕏@jassicsGitHub

ai-threat-model

Map prompts, data sources, tools, identities, and trust boundaries to identify AI-specific threats.

6|Updated May 30, 2026
Search Tags:#security#llm#threat-modeling#ai-security#genai
One-click install
npx skills add https://github.com/jassics/awesome-claude-security --skill ai-threat-model
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-threat-model
Source: https://github.com/jassics/awesome-claude-security/tree/main/plugins/llm-security/skills/ai-threat-model
Command: npx skills add https://github.com/jassics/awesome-claude-security --skill ai-threat-model

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill helps in threat modeling for LLM/RAG/agentic AI systems, identifying and mitigating AI-specific threats and trust boundary issues.

Core Features & Use Cases

  • AI-specific Threat Modeling: Maps prompts, data sources, tools, identities, and trust boundaries.
  • Threat Analysis: Enumerates AI-specific threats and mitigations.
  • Use Case: When designing or reviewing a GenAI feature's security, this Skill can help identify potential risks and implement necessary safeguards.

Quick Start

Run the ai-threat-model skill to create a threat model for your AI system.

Frequently Asked Questions about ai-threat-model

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

FAQPage Schema
How do I threat model a GenAI system to identify security risks?▼

Threat modeling a GenAI system requires mapping prompts, data sources, tools, identities, and trust boundaries to identify AI-specific threats and implement mitigations. It analyzes LLM, RAG, and agentic AI architectures to find vulnerabilities.

What AI-specific threats should I look for when securing LLM and RAG applications?▼

AI-specific threats in LLM and RAG applications target trust boundaries, prompt injections, and data source connections. Threat modeling identifies these risks by mapping identities and tools to enumerate vulnerabilities and suggest mitigations.

Does threat modeling for agentic AI require STRIDE analysis?▼

Threat modeling for agentic AI requires STRIDE analysis and security diagramming tools to effectively map trust boundaries and system components. These methodologies help identify threats across prompts, tools, and identities.

Can I use this approach to review the security of an existing GenAI feature?▼

Yes, you can use threat modeling to review an existing GenAI feature's security by mapping its data sources, prompts, and trust boundaries. This identifies potential risks and implements necessary safeguards.

What is the best way to map trust boundaries in an LLM application?▼

The best way to map trust boundaries in an LLM application is by diagramming prompts, data sources, tools, and identities. Threat modeling analyzes these boundaries to identify AI-specific risks and apply mitigations.

Related Skills

Explore More Software Engineering Skills→
KentoShimizuKentoShimizu
7

security-threat-modeling

Model assets, trust boundaries, and abuse paths to document security risks.

Community
Advanced
tense-itense-i
1

threat-mitigation-mapping

Map security threats to corresponding controls using Python dataclasses and enums.

Community
Advanced
bigeasyfreemanbigeasyfreeman
2

security-review

Automate STRIDE threat modeling and OWASP Top 10 scanning on code changes before merging.

Community
Advanced
haJ1thaJ1t

agent-shield

Monitor AI agent actions and detect security threats during runtime.

Community
Advanced