Fabric is a framework that puts AI at your fingertips. Instead of diving into chat interfaces (e.g., ChatGPT) or writing custom programs that consume APIs, you can create prompts as markdown text and receive output in markdown. Fabric also maintains a database of prompts called patterns.
In this repository, I have collected examples of the usage of the create_stride_threat_model pattern:
| File | Description | Model |
|---|---|---|
| INPUT.md | Example of an architecture document for the fictional project AI Nutrition-Pro | --- |
| baseline_threat_modeling_action.md | Baseline Threat Model from my ai-threat-modeling-action | Claude 3 Opus |
| baseline_threat_scenarios.md | Baseline Threat Model created by Fabric using the create_threat_scenarios pattern | Claude 3 Opus |
| baseline_stride_gpt.md | Baseline Threat Model created by STRIDE GPT | GPT-4o |
| threat_model_claude_3_opus.md | Threat Model created using the create_stride_threat_model pattern |
Claude 3 Opus |
| threat_model_gpt_4o.md | Threat Model created using the create_stride_threat_model pattern |
GPT-4o |
| threat_model_gemini_1.5_pro.md | Threat Model created using the create_stride_threat_model pattern |
Gemini-1.5 Pro Latest |
# Get Fabric installed - https://github.com/danielmiessler/fabric
$ wget https://raw.githubusercontent.com/xvnpw/fabric-stride-threat-model/main/INPUT.md
$ cat INPUT.md | fabric --pattern create_stride_threat_model -m claude-3-opus-20240229 > threat_model_claude_3_opus.md