Skip to content

Generate or Import a Threat Model

[!NOTE] Required license feature: threat

A threat model is produced one of two ways: generated by the Threat Modeling Agent (TMA), or imported from an existing OTM / OWASP Threat Dragon document.

Generating a model

  1. Click Generate — from the Threat Models list (choose a project in the dialog) or from a project's module list (the project is pre-selected).
  2. Pick a stage:
Stage What it produces
Run all (default) Builds the model, generates risks, then develops countermeasures.
Build model Inventory, system model and Data Flow Diagram only.
Generate risks STRIDE risks for an existing model.
Develop countermeasures Countermeasures for the existing risks.
  1. Confirm. The run starts in the background — it can take up to ten minutes — and a live progress panel shows the run status, elapsed time and, per phase, a status icon (running / completed / failed). A "View document" link appears once it's done, and you're notified if the run fails, times out, or the agent is already busy at its concurrency limit.

Every run is idempotent — running a stage again with nothing new to act on produces no new risks and leaves the document unchanged. See Run Modes for when to use each stage.

The agent reads the project's Application goals text (see Project Settings → General) to seed the model — populate that field, or upload an architecture.md, before generating for the best result.

Re-running a single stage

Open a model's detail page and use the Lifecycle dropdown: it lists each stage, shows whether it already has data, and lets you run or re-run just that stage without going back to the project.

Importing a model

Import an existing OTM or OWASP Threat Dragon document from the Threat Models list. It appears in the list alongside generated models, and opens in the same Data Flow Diagram viewer.

📸 Screenshot: the Generate dialog with the stage selector, and the live progress panel below it.