pentest-threat-model
Skill8 starsChanged 3 months agoMITView on GitHub
Threat modeling — STRIDE, DREAD, attack tree, data flow diagram, MITRE ATT&CK Navigator integration. Triggers on threat model, STRIDE, DREAD, attack tree, DFD, data flow diagram, MITRE ATT&CK, threat matrix, risk scoring.
In this repository63 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- ai-automationSkill
- consultingSkill
- contentSkill
- customer-successSkill
- data-analyticsSkill
- designSkill
- design-tokensSkill
- developmentSkill
- devopsSkill
- ecommerceSkill
- emailSkill
- expo-app-configSkill
- expo-config-pluginSkill
- expo-dev-clientSkill
- expo-eas-buildSkill
- expo-eas-submitSkill
- expo-eas-updateSkill
- expo-modulesSkill
- expo-notificationsSkill
- expo-orchestratorSkill
- expo-prebuildSkill
- expo-routerSkill
- expo-troubleshootingSkill
- financeSkill
- frontend-tasteSkill
- marketingSkill
- market-researchSkill
- mobileSkill
- pentest-adSkill
- pentest-apiSkill
- pentest-bizlogicSkill
- pentest-bugbountySkill
- pentest-cicdSkill
- pentest-cloudSkill
- pentest-credentialsSkill
- pentest-ctfSkill
- pentest-detectionSkill
- pentest-engagementSkill
- pentest-exploit-chainSkill
- pentest-forensicsSkill
- pentest-llmSkill
- pentest-malwareSkill
- pentest-mobileSkill
- pentest-opsec-evidenceSkill
- pentest-orchestratorSkill
- pentest-privescSkill
- pentest-reconSkill
- pentest-reportSkill
- pentest-socialSkill
- pentest-stigSkill
- pentest-webSkill
- pentest-wirelessSkill
- productivitySkill
- productSkill
- salesSkill
- security-checkSkill
- securitySkill
- seo-crawl-budgetSkill
- seoSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 154
When the user asks to persist the threat model (or when it feeds the security-check pipeline), write it to THREAT_MODEL.md at the scan …
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
pentest-threat-model
System threat modeling — STRIDE, DREAD, attack tree, DFD. Used at engagement start or in a design review.
Triggers
- "create a threat model"
- "STRIDE analysis"
- "DREAD scoring"
- "draw an attack tree"
- "data flow diagram"
- "MITRE ATT&CK matrix"
STRIDE (Microsoft)
| Letter | Threat | Counter Property |
| S | Spoofing identity | Authentication |
| T | Tampering with data | Integrity |
| R | Repudiation | Non-repudiation |
| I | Information disclosure | Confidentiality |
| D | Denial of service | Availability |
| E | Elevation of privilege | Authorization |
Apply STRIDE to every DFD element (Process, Data Store, Data Flow, Trust Boundary).
DREAD Scoring
| Dimension | 1 (Low) | 5 (High) | 10 (Critical) |
| Damage | Minimal | Major | Catastrophic |
| Reproducibility | Hard | Sometimes | Trivial |
| Exploitability | Hard | Moderate | Trivial (script) |
| Affected users | Few | Many | Most |
| Discoverability | Hidden | Apparent | Public docs |
Total: 50 max. High = high risk.
DFD Structure
[External Entity] --> ((Process)) --> [Data Store]
|
Trust Boundary
|
((Internal Service))Components:
- External Entity (square): user, 3rd party API
- Process (circle): web server, microservice
- Data Store (cylinder): database, cache
- Data Flow (arrow): request/response
- Trust Boundary (dashed line): VPC boundary, network zone
Attack Tree Template
Goal: Steal credit card data
├── 1. Compromise web server
│ ├── 1.1 RCE via SQL injection
│ │ └── 1.1.1 Find SQLi in /api
│ ├── 1.2 Steal session cookie via XSS
│ └── 1.3 Brute force admin panel
├── 2. Compromise database directly
│ ├── 2.1 Internal port reachable
│ └── 2.2 Default credential
└── 3. Social engineer admin
├── 3.1 Phishing
└── 3.2 Vishing IT helpdeskFor each leaf:
- DREAD score
- Likelihood (1-5)
- Cost to exploit (hours/days)
- Detection (likely/possible/unlikely)
MITRE ATT&CK Matrix Mapping
threat_model:
asset: Production database
threats:
- id: T-DB-001
stride: I (Information disclosure)
attack:
tactic: TA0009 Collection
technique: T1213 Data from Information Repositories
dread: { damage: 9, repro: 7, exploit: 6, affected: 8, discover: 5 }
score: 35
mitigation:
- DB encryption at rest
- Access logging + SIEM rule
- Network segmentationOutput Template
## Threat Model — <system>
### 1. System Description
- Frontend: React SPA on CDN
- Backend: Node.js API on Kubernetes
- Data: PostgreSQL (managed) + Redis (managed)
- 3rd party: Stripe, SendGrid, Auth0
### 2. Data Flow Diagram (Mermaid)
(mermaid flowchart of external entities, processes, data stores, and trust
boundaries — see the render example in the "Mermaid DFD Render" section below)
### 3. Entry Points & Trust Boundaries
Entry points: /api/* (JWT), /webhooks/stripe (signature), admin panel (SSO)
1. CDN <-> API (TLS, JWT auth)
2. API <-> DB (VPC private subnet, IAM auth)
3. API <-> 3rd party (TLS + API key)
### 4. Threats (STRIDE + DREAD table)
| # | Threat | STRIDE | DREAD | Tactic |
|---|--------|--------|-------|--------|
| 1 | Stripe webhook spoofing (no sig verify) | T | 38 | Initial Access |
| 2 | JWT alg=none bypass | S+E | 36 | Initial Access |
| 3 | DB credential leak via 0/1-day | I | 33 | Credential Access |
| 4 | Redis SSRF via SSRF in URL param | I | 30 | Discovery |
| 5 | DoS via excessive GraphQL nesting | D | 28 | Impact |
### 5. Mitigations
- Stripe: webhook signature verify (Stripe-Signature header)
- JWT: enforce alg whitelist (RS256 only)
- DB: rotate credentials quarterly, IAM auth (no static creds)
- SSRF: URL allowlist (no internal IP)
- GraphQL: max-depth 7, max-aliases 30File Output Convention (v1.34+)
When the user asks to persist the threat model (or when it feeds the security-check
pipeline), write it to THREAT_MODEL.md at the scan target's root (the project
root for a whole-repo scan) using the Output Template above, organized into numbered
sections:
- System Description
- Data Flow Diagram (Mermaid)
- Entry Points & Trust Boundaries
- Threats (STRIDE + DREAD table)
- Mitigations
Sections 3 and 4 are the machine-consumed scoping contract: the security-check
pipeline (Phase 1 recon and Phase 2 hunting) reads them to focus scanning on the
modeled entry points and threats. THREAT_MODEL.md is the head of the
harness-compatible artifact chain THREAT_MODEL.md → VULN-FINDINGS.json/.md →
TRIAGE.json/.md — filenames follow
anthropics/defending-code-reference-harness
(Apache-2.0) for interop. Unlike the generated scan artifacts, THREAT_MODEL.md
is a durable design document — commit it.
Mermaid DFD Render
flowchart LR
User[User Browser] -->|HTTPS| CDN[CDN]
CDN -->|HTTPS+JWT| API[API Gateway]
API -->|VPC| DB[(PostgreSQL)]
API -->|VPC| Cache[(Redis)]
API -->|HTTPS+API Key| Stripe[Stripe API]
API -->|HTTPS+API Key| SendGrid[SendGrid]
classDef boundary stroke-dasharray:5
style CDN stroke-dasharray:5
style API stroke-dasharray:5Out-of-Scope
- Live attack execution (modeling only)
- Automatically updating an existing model (manual review as needed)
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

