界面已本地化;技能介绍与说明文档按所选语言由 AI 自动翻译,代码、命令、路径和专业术语保留原文,以原作者文档为准。
探索技能
设计与创意 · SKILL.md

threat-modeling

threat-modeling

【设计与创意】Threat modeling workflow for software systems: scope, data flow diagrams, STRIDE analysis, risk scoring, and turning mitigations into backlog and tests. Use when designing new features, reviewing architecture changes, handling sensitive data, or hardening auth/payment/multi-tenant flows.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准

技能说明

GitHub

name: threat-modeling description: "Threat modeling workflow for software systems: scope, data flow diagrams, STRIDE analysis, risk scoring, and turning mitigations into backlog and tests. Use when designing new features, reviewing architecture changes, handling sensitive data, or hardening auth/payment/multi-tenant flows." user-invocable: false disable-model-invocation: true version: 1.0.0 category: universal author: Claude MPM Team license: MIT progressive_disclosure: entry_point: summary: "Run a lightweight threat modeling workshop (STRIDE) and turn risks into concrete mitigations, tests, and PR checks" when_to_use: "When designing new features, reviewing architecture changes, handling sensitive data, or hardening auth/payment/multi-tenant flows" quick_start: "1. Define scope/assets 2. Draw data flows + trust boundaries 3. STRIDE per element 4. Score + prioritize 5. Track mitigations + tests" token_estimate: entry: 150 full: 8000 context_limit: 900 tags:

  • security
  • threat-modeling
  • stride
  • architecture
  • risk requires_tools: []

Threat Modeling (STRIDE)

Workflow

  1. Scope — Identify the system boundary, assets (PII, credentials, payments), and availability requirements (SLO/SLA).
  2. Data Flow Diagram — Map actors, entry points, data stores, and external dependencies. Mark trust boundaries (public internet → edge → internal → database → third-party).
  3. 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.
  4. Risk score — Rate each threat by Impact (Low/Med/High) and Likelihood (Low/Med/High). Prioritize High-impact + Med/High-likelihood items first.
  5. Mitigate — Convert each prioritized threat into engineering tasks, verification tasks (tests, alerts), and operational controls (runbooks, access reviews).
  6. Tickets and tests — Create backlog items for mitigations and add abuse-case tests for critical flows. Add PR checklist items for ongoing verification.

Example: Threat Register Row

ElementSTRIDEThreatImpactLikelihoodMitigationOwnerStatus
API GatewaySpoofingStolen JWT reuse after session revocationHighMedShort-lived tokens (15 min TTL), refresh rotation, revocation list check on each requestSecurityOpen

This single row drives three artifacts: an engineering ticket (implement revocation-list middleware), a test (verify revoked token returns 401 within TTL window), and a PR checklist item (authz checks for new endpoints).

Validation Checkpoint

Before finalizing, verify completeness:

  • Every element in the data flow diagram has at least one STRIDE entry
  • All High-impact threats have an assigned owner and mitigation
  • Each mitigation maps to a backlog ticket or test case
  • Threat model doc includes assumptions and scope boundaries
  • PR checklist updated with new security requirements

Outputs (Definition of Done)

Produce a data flow diagram, a threat register, and a mitigation plan that becomes tickets and tests.

Load Next (References)

  • references/stride-workshop.md — step-by-step workshop agenda + DFD guidance
  • references/common-threats-and-mitigations.md — threat catalog with mitigations
  • references/templates.md — copy/paste templates for docs and tickets
我的收藏