Threat Modeling

Runs you through a lightweight STRIDE threat modeling workshop: you define scope and assets, map data flows with trust boundaries, apply STRIDE categories to each element, score risks, then convert the findings into backlog tickets and test cases. Built around three reference docs covering workshop steps, a threat catalog with ready-made mitigations, and templates you can drop straight into your issue tracker. The output is a data flow diagram, threat register, and mitigation plan. This is less "security theater" and more "turn architecture discussions into actual work items." Most useful when you're adding auth flows, handling PII, building multi-tenant features, or reviewing anything that crosses trust boundaries.

skills.sh installs: 474
GitHub stars: 79
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add bobmatnyc/claude-mpm-skills --skill threat-modeling --agent claude-code

Installs into .claude/skills of the current project.

Files

Threat Modeling (STRIDE)

Workflow

Scope — Identify the system boundary, assets (PII, credentials, payments), and availability requirements (SLO/SLA).
Data Flow Diagram — Map actors, entry points, data stores, and external dependencies. Mark trust boundaries (public internet → edge → internal → database → third-party).
STRIDE per element — For each element in the diagram, walk through all six STRIDE categories (Spoofing, Tampering, Repudiation, Information disclosure, Denial of service, Elevation of privilege) and record threats.
Risk score — Rate each threat by Impact (Low/Med/High) and Likelihood (Low/Med/High). Prioritize High-impact + Med/High-likelihood items first.
Mitigate — Convert each prioritized threat into engineering tasks, verification tasks (tests, alerts), and operational controls (runbooks, access reviews).
Tickets and tests — Create backlog items for mitigations and add abuse-case tests for critical flows. Add PR checklist items for ongoing verification.