Welcome to Threat Designer, an AI-powered agentic application for automated threat modeling. These guides will help you get started and make the most of Threat Designer's capabilities.
New to Threat Designer? Start here to understand the complete workflow from submission to analysis.
Learn how to create your first threat model by uploading an architecture diagram and configuring AI analysis parameters. This guide covers:
- Completing the submission form with required and optional fields
- Choosing the right iteration count and reasoning boost settings
- Understanding processing times and what to expect
- Tips for creating effective architecture diagrams
Start here if: You're submitting your first threat model or want to understand the submission options better.
Once your threat model is complete, these guides help you understand, refine, and enhance your analysis.
Navigate and modify your completed threat model results. This guide covers:
- Understanding the STRIDE-based structure (Assets, Flows, Trust Boundaries, Threat Sources, Threat Catalog)
- Editing, adding, and deleting entries across all sections
- Saving your changes and avoiding data loss
- Downloading threat models in multiple formats (PDF, DOCX, JSON)
Start here if: Your threat model has finished processing and you want to review or customize the results.
Re-run AI analysis on existing threat models to incorporate changes and generate updated insights. This guide covers:
- When and why to use replay instead of creating a new threat model
- Preserving important threats with the starring system
- Adjusting analysis parameters and adding focused instructions
- Understanding what gets preserved vs. updated during replay
Start here if: You've made manual edits to your threat model or want to refine the analysis with different parameters.
Visualize and analyze attack paths with interactive attack tree diagrams. This guide covers:
- Understanding attack tree structure (goals, gates, leaf attacks, countermeasures)
- Generating attack trees for specific threats
- Using focus mode to analyze complex attack branches
- Interpreting AND/OR logic gates to identify defensive opportunities
Start here if: You want to understand how specific threats could be executed and identify the most effective defensive measures.
Interact with Sentry, Threat Designer's built-in AI assistant, for conversational threat analysis. This guide covers:
- Asking Sentry to identify gaps and missing threats
- Getting mitigation recommendations and security guidance
- Having Sentry directly modify your threat catalog through conversation
- Leveraging AWS-specific knowledge for cloud architectures
Start here if: You want to explore your threat model interactively, ask security questions, or get AI-assisted recommendations.
Attach a personal knowledge base to your threat models to surface organization-specific context. This guide covers:
- Creating a Space and uploading documents (runbooks, architecture docs, security policies)
- Attaching a Space when submitting a threat model
- Understanding how the agent queries your Space and injects insights
- Interpreting the Context tab in the analysis trail
Start here if: You want the agent to incorporate your own documentation—compliance requirements, existing controls, data classification policies—into the threat modeling analysis.
Create new versions of existing threat models when your architecture changes. This guide covers:
- Understanding when to use versioning vs. replay
- Uploading a new architecture diagram and configuring version options
- How the AI agent diffs architectures and incrementally updates assets, flows, boundaries, and threats
- Monitoring progress and reviewing versioned results
Start here if: Your architecture has changed and you want to update your threat model incrementally without starting from scratch.
Share threat models with team members and work together while maintaining data integrity. This guide covers:
- Sharing threat models and managing collaborator access levels
- Understanding edit locks and how they prevent conflicts
- Resolving conflicts when multiple users make changes
Start here if: You want to work on threat models with your team or need to share results with stakeholders.
For the best experience with Threat Designer, we recommend this workflow:
- Submit → Create your initial threat model with an architecture diagram (Submission Guide)
- Review → Examine the AI-generated results and make manual refinements (Interaction Guide)
- Visualize → Generate attack trees for critical threats to understand attack paths (Attack Tree Guide)
- Collaborate → Share with team members for review and collective input (Collaboration Guide)
- Enhance → Use Sentry to explore gaps, improve descriptions, and add missing threats (Sentry Guide)
- Refine → Replay with updated parameters or instructions to expand your analysis (Replay Guide)
- Version → When your architecture changes, create a new version to incrementally update the analysis (Versioning Guide)
- Iterate → Repeat steps 2-7 as your architecture evolves or your understanding deepens
| Task | Guide | Time Required |
|---|---|---|
| Create first threat model | Submit | 5 min setup + 15-30 min processing |
| Review and edit results | Interact | 10-30 min |
| Visualize attack paths | Attack Trees | 30 sec per threat |
| Attach org knowledge to threat models | Spaces | 5-10 min setup |
| Share with team members | Collaborate | 2-5 min |
| Ask AI questions about threats | Sentry | Ongoing |
| Update with new parameters | Replay | 5 min setup + 5-20 min processing |
| Update after architecture changes | Versioning | 5 min setup + 5-15 min processing |
Each guide includes:
- Step-by-step instructions with screenshots where helpful
- Best practices and tips for success
- Common use cases and examples
Remember: Threat modeling is an iterative process. Don't expect perfection on your first submission—use Threat Designer's tools (manual editing, Sentry, and Replay) to continuously refine and improve your security analysis.